)]}' { "log": [ { "commit": "45196cee28a5bcfb6ddbe2bffa4270cbed66ae4b", "tree": "1ee61842ac03e05f5fa346b7e2095c7fd80ceabd", "parents": [ "e2d4370b78f52d78e21695abd610a7044268763c", "c6c1e4491dc8d1ed2509fa6aacffa7f34614fc38" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 22 13:00:53 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 22 13:00:53 2012 -0800" }, "message": "Merge tag \u0027usb-3.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nUSB bugfixes for 3.3-rc4\n\nA number of new device ids, and a cleanup/fix for some of the option\ndevice ids that shouldn\u0027t have been added in the first place.\n\nThere\u0027s also a few USB 3 fixes for problems that people have reported,\nand a usb-storage bugfix to round it out.\n\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\n* tag \u0027usb-3.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:\n USB: Added Kamstrup VID/PIDs to cp210x serial driver.\n USB: Serial: ti_usb_3410_5052: Add Abbot Diabetes Care cable id\n usb-storage: fix freezing of the scanning thread\n xhci: Fix encoding for HS bulk/control NAK rate.\n USB: Set hub depth after USB3 hub reset\n USB: Fix handoff when BIOS disables host PCI device.\n USB: option: cleanup zte 3g-dongle\u0027s pid in option.c\n USB: Don\u0027t fail USB3 probe on missing legacy PCI IRQ.\n xhci: Fix oops caused by more USB2 ports than USB3 ports.\n USB: Remove duplicate USB 3.0 hub feature #defines.\n" }, { "commit": "e2d4370b78f52d78e21695abd610a7044268763c", "tree": "e11a05237a7c71c7c8caa9e9ceb0708bfa39ea49", "parents": [ "71c01b9d5b00e5887f6a81a5303c65841e096d61", "3ac0eb6d62fde0a60a6c5c61e562af1db8fbf712" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 22 11:58:30 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 22 11:58:30 2012 -0800" }, "message": "Merge branch \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux\n\nIntel, radeon, exynos fixes.\n\nIntel: fixes a few Ivybridge hangs, along with fixing RC6 on SNB (still\nnot on, but at least allows for distros to patch it on easily).\n\nradeon: oops reading some files in debugfs that weren\u0027t meant to appear,\na fix that touches a lot of files, so looks worse than it is, it fixes\nan oops if a GPU reset fails and userspace keeps submitting more data,\nalong with a minor BIOS fix for newer boards.\n\nexynos: a group of fixes for exynos, they\u0027ve sent me a few more but\nthese were all I got through, and its no hw vanilla kernel users see a\nlot off yet.\n\n* \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux:\n drm/radeon/kms/atom: dpms bios scratch reg updates\n drm/radeon/kms: properly set accel working flag and bailout when false\n drm/radeon: Only create additional ring debugfs files on Cayman or newer.\n drm/exynos: added postclose to release resource.\n drm/exynos: removed exynos_drm_fbdev_recreate function.\n drm/exynos: fixed page flip issue.\n drm/exynos: added possible_clones setup function.\n drm/exynos: removed pageflip_event_list init code when closed.\n drm/exynos: changed priority of mixer layers.\n drm/exynos: Fix typo in exynos_mixer.c\n drm/i915: do not enable RC6p on Sandy Bridge\n drm/i915: gen7: Disable the RHWO optimization as it can cause GPU hangs.\n drm/i915: gen7: work around a system hang on IVB\n drm/i915: gen7: Implement an L3 caching workaround.\n drm/i915: gen7: implement rczunit workaround\n" }, { "commit": "71c01b9d5b00e5887f6a81a5303c65841e096d61", "tree": "fd9420a67e0a7c7c39361622dc704b57bfc8f17e", "parents": [ "437cf4c7b7feeca29f74e1e6f8d458ea2623d5e2", "3372f5a7d005dd42e754490fed6a0171c4a018c6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 22 08:45:08 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 22 08:45:08 2012 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu\n\nIt contains 3 important fixes for ColdFire based machines:\n - fix processes getting stuck when running from strace\n - fix kernel vmalloced pages not being visible in all kernel contexts\n - fix shared user pages sometimes being visible in another process\n context\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu:\n m68k: Do not set global share for non-kernel shared pages\n m68k: Add shared bit to Coldfire kernel page entries\n m68knommu: fix syscall tracing stuck process\n" }, { "commit": "437cf4c7b7feeca29f74e1e6f8d458ea2623d5e2", "tree": "71ef02369cbbef4523a0cb426809e724b85b723a", "parents": [ "719741d9986572d64b47c35c09f5e7bb8d389400", "abe9a6d57b4544ac208401f9c0a4262814db2be4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 22 08:43:35 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 22 08:43:35 2012 -0800" }, "message": "Merge tag \u0027nfs-for-3.3-4\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs\n\nBugfixes for the NFS client.\n\nFix a nasty Oops in the NFSv4 getacl code, another source of infinite\nloops in the NFSv4 state recovery code, and a regression in NFSv4.1\nsession initialisation.\n\nAlso deal with an NFSv4.1 memory leak.\n\n* tag \u0027nfs-for-3.3-4\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs:\n NFSv4: fix server_scope memory leak\n NFSv4.1: Fix a NFSv4.1 session initialisation regression\n NFSv4: Ensure we throw out bad delegation stateids on NFS4ERR_BAD_STATEID\n NFSv4: Fix an Oops in the NFSv4 getacl code\n" }, { "commit": "3ac0eb6d62fde0a60a6c5c61e562af1db8fbf712", "tree": "d1a934fc3283be67df7eb4c6aa4adaca6d890b10", "parents": [ "6b7746e8768e1c550b320d5af761f73e5aa37f76" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Sun Feb 19 21:42:03 2012 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Feb 22 10:30:06 2012 +0000" }, "message": "drm/radeon/kms/atom: dpms bios scratch reg updates\n\ndpms bits not used on DCE4+\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "6b7746e8768e1c550b320d5af761f73e5aa37f76", "tree": "ad04aeca4bbc904418d21717f17689f1c0aa5bf7", "parents": [ "f0d14daa6906070ca044b86f483fdde7d81f5294" ], "author": { "name": "Jerome Glisse", "email": "jglisse@redhat.com", "time": "Mon Feb 20 17:57:20 2012 -0500" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Feb 22 10:30:02 2012 +0000" }, "message": "drm/radeon/kms: properly set accel working flag and bailout when false\n\nIf accel is not working many subsystem such as the ib pool might not be\ninitialized properly that can lead to segfault inside kernel when cs\nioctl is call with non working acceleration. To avoid this make sure\nthe accel working flag is false when an error in GPU startup happen and\nreturn EBUSY from cs ioctl if accel is not working.\n\nSigned-off-by: Jerome Glisse \u003cjglisse@redhat.com\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "f0d14daa6906070ca044b86f483fdde7d81f5294", "tree": "20c123098eeeeec8f2161e6b508343610f652d90", "parents": [ "bb757a7e251f73ce6626689f8be4bb8ba86933cd" ], "author": { "name": "Michel Dänzer", "email": "michel.daenzer@amd.com", "time": "Tue Feb 21 17:39:15 2012 +0100" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Feb 22 10:28:52 2012 +0000" }, "message": "drm/radeon: Only create additional ring debugfs files on Cayman or newer.\n\nBugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d46274\n\nTested with a Cayman card in a Llano system: The additional files are created\nand working for the Cayman card but not created for the CPU\u0027s built-in GPU.\n\nSigned-off-by: Michel Dänzer \u003cmichel.daenzer@amd.com\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "bb757a7e251f73ce6626689f8be4bb8ba86933cd", "tree": "17cc7b2858c4d8ddaf6acd94ab45b033f450f4dd", "parents": [ "53ef299f3900bc1deb163b94d4f1cac4f3346152", "1c8ecf80fdee4e7b23a9e7da7ff9bd59ba2dcf96" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Feb 22 08:02:17 2012 +0000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Feb 22 08:02:17 2012 +0000" }, "message": "Merge branch \u0027drm-intel-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/drm-intel into drm-fixes\n\n* \u0027drm-intel-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/drm-intel:\n drm/i915: do not enable RC6p on Sandy Bridge\n drm/i915: gen7: Disable the RHWO optimization as it can cause GPU hangs.\n drm/i915: gen7: work around a system hang on IVB\n drm/i915: gen7: Implement an L3 caching workaround.\n drm/i915: gen7: implement rczunit workaround\n" }, { "commit": "719741d9986572d64b47c35c09f5e7bb8d389400", "tree": "c8cb9001989debd328711a4cd19f7a7cb7877dfe", "parents": [ "6b0d1abb354d2d6b991c051e7380bd484a05e61f", "9b45c0d2c50e19bd73d2f5e18dd6912a6991306e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 21 18:25:30 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 21 18:25:30 2012 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:\n maintainers: update my email address\n" }, { "commit": "6b0d1abb354d2d6b991c051e7380bd484a05e61f", "tree": "e39663a9a4b949de7024af9780f716e16377960d", "parents": [ "faf309009e2e18d30c032b7d9479f29b91677c37", "5180bb392a8aab5233e6db858ac1d8371533e20f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 21 18:24:42 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 21 18:24:42 2012 -0800" }, "message": "Merge branch \u0027fixes\u0027 of git://git.linaro.org/people/rmk/linux-arm\n\nA few more things this time around. The only thing warranting some\ncommentry is the modpost change, which allows folk building a Thumb2\nenabled kernel to see section mismatch warnings. This is why many\nweren\u0027t noticed with OMAP.\n\n* \u0027fixes\u0027 of git://git.linaro.org/people/rmk/linux-arm:\n ARM/audit: include audit header and fix audit arch\n ARM: OMAP: fix voltage domain build errors with PM_OPP disabled\n ARM/PCI: Remove ARM\u0027s duplicate definition of \u0027pcibios_max_latency\u0027\n ARM: 7336/1: smp_twd: Don\u0027t register CPUFREQ notifiers if local timers are not initialised\n ARM: 7327/1: need to include asm/system.h in asm/processor.h\n ARM: 7326/2: PL330: fix null pointer dereference in pl330_chan_ctrl()\n ARM: 7164/3: PL330: Fix the size of the dst_cache_ctrl field\n ARM: 7325/1: fix v7 boot with lockdep enabled\n ARM: 7324/1: modpost: Fix section warnings for ARM for many compilers\n ARM: 7323/1: Do not allow ARM_LPAE on pre-ARMv7 architectures\n" }, { "commit": "9b45c0d2c50e19bd73d2f5e18dd6912a6991306e", "tree": "840aa31e544d5ff5db1c99333a39a66f20578f25", "parents": [ "27e74da9800289e69ba907777df1e2085231eff7" ], "author": { "name": "James Morris", "email": "james.l.morris@oracle.com", "time": "Wed Feb 22 12:45:07 2012 +1100" }, "committer": { "name": "James Morris", "email": "james.l.morris@oracle.com", "time": "Wed Feb 22 12:45:07 2012 +1100" }, "message": "maintainers: update my email address\n\nUpdate my email address.\n\nSigned-off-by: James Morris \u003cjames.l.morris@oracle.com\u003e\n" }, { "commit": "faf309009e2e18d30c032b7d9479f29b91677c37", "tree": "09a22833eaf02886cc1de6ac513aad1143dcf822", "parents": [ "797a796a13df6b84a4791e57306737059b5b2384" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 21 17:24:20 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 21 17:24:20 2012 -0800" }, "message": "sys_poll: fix incorrect type for \u0027timeout\u0027 parameter\n\nThe \u0027poll()\u0027 system call timeout parameter is supposed to be \u0027int\u0027, not\n\u0027long\u0027.\n\nNow, the reason this matters is that right now 32-bit compat mode is\nbroken on at least x86-64, because the 32-bit code just calls\n\u0027sys_poll()\u0027 directly on x86-64, and the 32-bit argument will have been\nzero-extended, turning a signed \u0027int\u0027 into a large unsigned \u0027long\u0027\nvalue.\n\nWe could just introduce a \u0027compat_sys_poll()\u0027 function for this, and\nthat may eventually be what we have to do, but since the actual standard\npoll() semantics is *supposed* to be \u0027int\u0027, and since at least on x86-64\nglibc sign-extends the argument before invocing the system call (so\nnobody can actually use a 64-bit timeout value in user space _anyway_,\neven in 64-bit binaries), the simpler solution would seem to be to just\nfix the definition of the system call to match what it should have been\nfrom the very start.\n\nIf it turns out that somebody somehow circumvents the user-level libc\n64-bit sign extension and actually uses a large unsigned 64-bit timeout\ndespite that not being how poll() is supposed to work, we will need to\ndo the compat_sys_poll() approach.\n\nReported-by: Thomas Meyer \u003cthomas@m3y3r.de\u003e\nAcked-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "797a796a13df6b84a4791e57306737059b5b2384", "tree": "400d13deef98fdf3fc893e9e55f208c5716b8ba9", "parents": [ "27e74da9800289e69ba907777df1e2085231eff7" ], "author": { "name": "Hitoshi Mitake", "email": "mitake@dcl.info.waseda.ac.jp", "time": "Tue Feb 07 11:45:33 2012 +0900" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 21 16:47:28 2012 -0800" }, "message": "asm-generic: architecture independent readq/writeq for 32bit environment\n\nThis provides unified readq()/writeq() helper functions for 32-bit\ndrivers.\n\nFor some cases, readq/writeq without atomicity is harmful, and order of\nio access has to be specified explicitly. So in this patch, new two\nheader files which contain non-atomic readq/writeq are added.\n\n - \u003casm-generic/io-64-nonatomic-lo-hi.h\u003e provides non-atomic readq/\n writeq with the order of lower address -\u003e higher address\n\n - \u003casm-generic/io-64-nonatomic-hi-lo.h\u003e provides non-atomic readq/\n writeq with reversed order\n\nThis allows us to remove some readq()s that were added drivers when the\ndefault non-atomic ones were removed in commit dbee8a0affd5 (\"x86:\nremove 32-bit versions of readq()/writeq()\")\n\nThe drivers which need readq/writeq but can do with the non-atomic ones\nmust add the line:\n\n #include \u003casm-generic/io-64-nonatomic-lo-hi.h\u003e /* or hi-lo.h */\n\nBut this will be nop in 64-bit environments, and no other #ifdefs are\nrequired. So I believe that this patch can solve the problem of\n 1. driver-specific readq/writeq\n 2. atomicity and order of io access\n\nThis patch is tested with building allyesconfig and allmodconfig as\nARCH\u003dx86 and ARCH\u003di386 on top of tip/master.\n\nCc: Kashyap Desai \u003cKashyap.Desai@lsi.com\u003e\nCc: Len Brown \u003clenb@kernel.org\u003e\nCc: Ravi Anand \u003cravi.anand@qlogic.com\u003e\nCc: Vikas Chaudhary \u003cvikas.chaudhary@qlogic.com\u003e\nCc: Matthew Garrett \u003cmjg@redhat.com\u003e\nCc: Jason Uhlenkott \u003cjuhlenko@akamai.com\u003e\nCc: James Bottomley \u003cJames.Bottomley@parallels.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\nCc: Roland Dreier \u003croland@purestorage.com\u003e\nCc: James Bottomley \u003cjbottomley@parallels.com\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nCc: Matthew Wilcox \u003cmatthew.r.wilcox@intel.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Hitoshi Mitake \u003ch.mitake@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c6c1e4491dc8d1ed2509fa6aacffa7f34614fc38", "tree": "b3937360eeb793dcfb30149f5ec02f1b70001433", "parents": [ "7fd25702ba616d9ba56e2a625472f29e5aff25ee" ], "author": { "name": "Bruno Thomsen", "email": "bruno.thomsen@gmail.com", "time": "Tue Feb 21 23:41:37 2012 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Feb 21 16:29:15 2012 -0800" }, "message": "USB: Added Kamstrup VID/PIDs to cp210x serial driver.\n\nSigned-off-by: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "7fd25702ba616d9ba56e2a625472f29e5aff25ee", "tree": "cbd75547c475501a0bc8404f11501dffbfa207f3", "parents": [ "bb94a406682770a35305daaa241ccdb7cab399de" ], "author": { "name": "Andrew Lunn", "email": "andrew@lunn.ch", "time": "Mon Feb 20 09:31:57 2012 +0100" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Feb 21 16:29:15 2012 -0800" }, "message": "USB: Serial: ti_usb_3410_5052: Add Abbot Diabetes Care cable id\n\nThis USB-serial cable with mini stereo jack enumerates as:\nBus 001 Device 004: ID 1a61:3410 Abbott Diabetes Care\n\nIt is a TI3410 inside.\n\nSigned-off-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "bb94a406682770a35305daaa241ccdb7cab399de", "tree": "a6801c8edec8809e34c34ec889db4682f074c923", "parents": [ "9a9a71b77c3fd511e5dda6236deb8a02d156b864" ], "author": { "name": "Alan Stern", "email": "stern@rowland.harvard.edu", "time": "Tue Feb 21 13:16:32 2012 -0500" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Feb 21 16:29:15 2012 -0800" }, "message": "usb-storage: fix freezing of the scanning thread\n\nThis patch (as1521b) fixes the interaction between usb-storage\u0027s\nscanning thread and the freezer. The current implementation has a\nrace: If the device is unplugged shortly after being plugged in and\njust as a system sleep begins, the scanning thread may get frozen\nbefore the khubd task. Khubd won\u0027t be able to freeze until the\ndisconnect processing is complete, and the disconnect processing can\u0027t\nproceed until the scanning thread finishes, so the sleep transition\nwill fail.\n\nThe implementation in the 3.2 kernel suffers from an additional\nproblem. There the scanning thread calls set_freezable_with_signal(),\nand the signals sent by the freezer will mess up the thread\u0027s I/O\ndelays, which are all interruptible.\n\nThe solution to both problems is the same: Replace the kernel thread\nused for scanning with a delayed-work routine on the system freezable\nwork queue. Freezable work queues have the nice property that you can\ncancel a work item even while the work queue is frozen, and no signals\nare needed.\n\nThe 3.2 version of this patch solves the problem in Bugzilla #42730.\n\nSigned-off-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\nAcked-by: Seth Forshee \u003cseth.forshee@canonical.com\u003e\nCC: stable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "9a9a71b77c3fd511e5dda6236deb8a02d156b864", "tree": "304eb27f688f4420ce82b5689a1f1e74aab4ae0f", "parents": [ "b9e44fe5ecda4158c22bc1ea4bffa378a4f83f65", "340a3504fd39dad753ba908fb6f894ee81fc3ae2" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Feb 21 16:25:30 2012 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Feb 21 16:25:30 2012 -0800" }, "message": "Merge tag \u0027for-usb-linus-2012-02-21\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/sarah/xhci into usb-linus\n\nHi Greg,\n\nHere\u0027s three bug fixes that should be queued for 3.3.\n\nThe first fixes an issue we saw with an Intel Panther Point xHCI host,\nwhere a certain OSV\u0027s custom BIOS would disable the PCI device during\nboot. It changes the generic PCI quirks handler for all USB host\ncontrollers, but in a way both Jesse Barnes and Oliver Neukum have\nagreed is safe.\n\nThe second patch is Elric Fu\u0027s first kernel patch! Congrats! It fixes\na bug in the USB 3.0 hub reset handling.\n\nThe last patch fixes a bug in the xHCI driver that feeds invalid input\nto the xHC host. Only the VIA host controller seems to have issues with\nit. Thanks to Felipe Contreras for testing this patch on his VIA host,\nand Andiry Xu for suggesting the fix.\n\nAll three patches are marked for stable.\n\nSarah Sharp\n" }, { "commit": "340a3504fd39dad753ba908fb6f894ee81fc3ae2", "tree": "304eb27f688f4420ce82b5689a1f1e74aab4ae0f", "parents": [ "a45aa3b30583e7d54e7cf4fbcd0aa699348a6e5c" ], "author": { "name": "Sarah Sharp", "email": "sarah.a.sharp@linux.intel.com", "time": "Mon Feb 13 14:42:11 2012 -0800" }, "committer": { "name": "Sarah Sharp", "email": "sarah.a.sharp@linux.intel.com", "time": "Tue Feb 21 15:48:46 2012 -0800" }, "message": "xhci: Fix encoding for HS bulk/control NAK rate.\n\nThe xHCI 0.96 spec says that HS bulk and control endpoint NAK rate must\nbe encoded as an exponent of two number of microframes. The endpoint\ndescriptor has the NAK rate encoded in number of microframes. We were\njust copying the value from the endpoint descriptor into the endpoint\ncontext interval field, which was not correct. This lead to the VIA\nhost rejecting the add of a bulk OUT endpoint from any USB 2.0 mass\nstorage device.\n\nThe fix is to use the correct encoding. Refactor the code to convert\nnumber of frames to an exponential number of microframes, and make sure\nwe convert the number of microframes in HS bulk and control endpoints to\nan exponent.\n\nThis should be back ported to kernels as old as 2.6.31, that contain the\ncommit dfa49c4ad120a784ef1ff0717168aa79f55a483a \"USB: xhci - fix math\nin xhci_get_endpoint_interval\"\n\nSigned-off-by: Sarah Sharp \u003csarah.a.sharp@linux.intel.com\u003e\nTested-by: Felipe Contreras \u003cfelipe.contreras@gmail.com\u003e\nSuggested-by: Andiry Xu \u003candiry.xu@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "a45aa3b30583e7d54e7cf4fbcd0aa699348a6e5c", "tree": "98a28b1d77d0425000a77f55868bc5367065107c", "parents": [ "cab928ee1f221c9cc48d6615070fefe2e444384a" ], "author": { "name": "Elric Fu", "email": "elricfu1@gmail.com", "time": "Sat Feb 18 13:32:27 2012 +0800" }, "committer": { "name": "Sarah Sharp", "email": "sarah.a.sharp@linux.intel.com", "time": "Tue Feb 21 15:45:25 2012 -0800" }, "message": "USB: Set hub depth after USB3 hub reset\n\nThe superspeed device attached to a USB 3.0 hub(such as VIA\u0027s)\ndoesn\u0027t respond the address device command after resume. The\nroot cause is the superspeed hub will miss the Hub Depth value\nthat is used as an offset into the route string to locate the\nbits it uses to determine the downstream port number after\nreset, and all packets can\u0027t be routed to the device attached\nto the superspeed hub.\n\nHub driver sends a Set Hub Depth request to the superspeed hub\nexcept for USB 3.0 root hub when the hub is initialized and\ndoesn\u0027t send the request again after reset due to the resume\nprocess. So moving the code that sends the Set Hub Depth request\nto the superspeed hub from hub_configure() to hub_activate()\nis to cover those situations include initialization and reset.\n\nThe patch should be backported to kernels as old as 2.6.39.\n\nSigned-off-by: Elric Fu \u003celricfu1@gmail.com\u003e\nSigned-off-by: Sarah Sharp \u003csarah.a.sharp@linux.intel.com\u003e\nAcked-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "cab928ee1f221c9cc48d6615070fefe2e444384a", "tree": "80922aa70a57cf99fb94df909beacfe56e6f2c0e", "parents": [ "b9e44fe5ecda4158c22bc1ea4bffa378a4f83f65" ], "author": { "name": "Sarah Sharp", "email": "sarah.a.sharp@linux.intel.com", "time": "Tue Feb 07 15:11:46 2012 -0800" }, "committer": { "name": "Sarah Sharp", "email": "sarah.a.sharp@linux.intel.com", "time": "Tue Feb 21 15:45:08 2012 -0800" }, "message": "USB: Fix handoff when BIOS disables host PCI device.\n\nOn some systems with an Intel Panther Point xHCI host controller, the\nBIOS disables the xHCI PCI device during boot, and switches the xHCI\nports over to EHCI. This allows the BIOS to access USB devices without\nhaving xHCI support.\n\nThe downside is that the xHCI BIOS handoff mechanism will fail because\nmemory mapped I/O is not enabled for the disabled PCI device.\nJesse Barnes says this is expected behavior. The PCI core will enable\nBARs before quirks run, but it will leave it in an undefined state, and\nit may not have memory mapped I/O enabled.\n\nMake the generic USB quirk handler call pci_enable_device() to re-enable\nMMIO, and call pci_disable_device() once the host-specific BIOS handoff\nis finished. This will balance the ref counts in the PCI core. When\nthe PCI probe function is called, usb_hcd_pci_probe() will call\npci_enable_device() again.\n\nThis should be back ported to kernels as old as 2.6.31. That was the\nfirst kernel with xHCI support, and no one has complained about BIOS\nhandoffs failing due to memory mapped I/O being disabled on other hosts\n(EHCI, UHCI, or OHCI).\n\nSigned-off-by: Sarah Sharp \u003csarah.a.sharp@linux.intel.com\u003e\nAcked-by: Oliver Neukum \u003coneukum@suse.de\u003e\nCc: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "5180bb392a8aab5233e6db858ac1d8371533e20f", "tree": "8ce5c857bdca8abfefcaadc05843e3c0dbea87ec", "parents": [ "3ddd4d0c629f3d013aa1ab24bb0546dfe342ff94" ], "author": { "name": "Eric Paris", "email": "eparis@redhat.com", "time": "Tue Feb 21 11:26:55 2012 -0500" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Tue Feb 21 16:50:14 2012 +0000" }, "message": "ARM/audit: include audit header and fix audit arch\n\nBoth bugs being fixed were introduced in:\n29ef73b7a823b77a7cd0bdd7d7cded3fb6c2587b\n\nInclude linux/audit.h to fix below build errors:\n\n CC arch/arm/kernel/ptrace.o\narch/arm/kernel/ptrace.c: In function \u0027syscall_trace\u0027:\narch/arm/kernel/ptrace.c:919: error: implicit declaration of function \u0027audit_syscall_exit\u0027\narch/arm/kernel/ptrace.c:921: error: implicit declaration of function \u0027audit_syscall_entry\u0027\narch/arm/kernel/ptrace.c:921: error: \u0027AUDIT_ARCH_ARMEB\u0027 undeclared (first use in this function)\narch/arm/kernel/ptrace.c:921: error: (Each undeclared identifier is reported only once\narch/arm/kernel/ptrace.c:921: error: for each function it appears in.)\nmake[1]: *** [arch/arm/kernel/ptrace.o] Error 1\nmake: *** [arch/arm/kernel] Error 2\n\nThis part of the patch is:\nReported-by: Axel Lin \u003caxel.lin@gmail.com\u003e\nReported-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n(They both provided patches to fix it)\n\nThis patch also (at the request of the list) fixes the fact that\nARM has both LE and BE versions however the audit code was called as if\nit was always BE. If audit userspace were to try to interpret the bits\nit got from a LE system it would obviously do so incorrectly. Fix this\nby using the right arch flag on the right system.\n\nThis part of the patch is:\nReported-by: Russell King - ARM Linux \u003clinux@arm.linux.org.uk\u003e\n\nSigned-off-by: Eric Paris \u003ceparis@redhat.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "3ddd4d0c629f3d013aa1ab24bb0546dfe342ff94", "tree": "ce83fad9ef10554af0e1f69cd6d9a3f9dfd9e090", "parents": [ "e23e8c0690d2952dce53e712d01d5b2179f98b64" ], "author": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Feb 15 11:28:06 2012 +0000" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Tue Feb 21 09:36:34 2012 +0000" }, "message": "ARM: OMAP: fix voltage domain build errors with PM_OPP disabled\n\nThe voltage domain code wants the voltage tables, which are in the\nopp*.c files. These files aren\u0027t built when PM_OPP is disabled,\ncausing the following build errors at link time:\n\ntwl-common.c:(.init.text+0x2e48): undefined reference to `omap34xx_vddmpu_volt_data\u0027\ntwl-common.c:(.init.text+0x2e4c): undefined reference to `omap34xx_vddcore_volt_data\u0027\ntwl-common.c:(.init.text+0x2e5c): undefined reference to `omap36xx_vddmpu_volt_data\u0027\ntwl-common.c:(.init.text+0x2e60): undefined reference to `omap36xx_vddcore_volt_data\u0027\ntwl-common.c:(.init.text+0x2830): undefined reference to `omap44xx_vdd_mpu_volt_data\u0027\ntwl-common.c:(.init.text+0x283c): undefined reference to `omap44xx_vdd_iva_volt_data\u0027\ntwl-common.c:(.init.text+0x2844): undefined reference to `omap44xx_vdd_core_volt_data\u0027\n\nAcked-by: Kevin Hilman \u003ckhilman@ti.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "e23e8c0690d2952dce53e712d01d5b2179f98b64", "tree": "df2533028276a094b2b38c5922b0cf6561e5d1eb", "parents": [ "910ba598c818243cf3de2f97e03c49948bbb0511" ], "author": { "name": "Myron Stowe", "email": "mstowe@redhat.com", "time": "Tue Feb 07 15:26:44 2012 -0700" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Tue Feb 21 09:35:32 2012 +0000" }, "message": "ARM/PCI: Remove ARM\u0027s duplicate definition of \u0027pcibios_max_latency\u0027\n\nThe patch series to re-factor PCI\u0027s \u0027latency timer\u0027 setup (re:\nhttp://marc.info/?l\u003dlinux-kernel\u0026m\u003d131983853831049\u0026w\u003d2) forgot to\nremove the ARM specific definition of \u0027pcibios_max_latency\u0027 once such\nhad been moved into the pci core resulting in ARM related compile\nerrors -\n drivers/built-in.o:(.data+0x230): multiple definition of\n `pcibios_max_latency\u0027\n arch/arm/common/built-in.o:(.data+0x40c): first defined here\n make[1]: *** [vmlinux.o] Error 1\n\nIn the series, patch 2/16 (commit 168c8619fd8) converted the ARM\nspecific version of \u0027pcibios_set_master()\u0027 to a non-inlined version.\nThis was done in preperation for hosting it up into PCI\u0027s core, which\nwas done in patch 10/16 (commit 96c5590058d) of the series (and\nwhere the removal of ARM\u0027s \u0027pcibios_max_latency\u0027 was overlooked).\n\nReported-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: Myron Stowe \u003cmyron.stowe@redhat.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "910ba598c818243cf3de2f97e03c49948bbb0511", "tree": "b8508a68eafcda631c2ce624b418e46c69a00282", "parents": [ "fee6a3c33a8f137f39cd9997b0476411f73576c7" ], "author": { "name": "Santosh Shilimkar", "email": "santosh.shilimkar@ti.com", "time": "Tue Feb 21 10:24:22 2012 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Tue Feb 21 09:26:46 2012 +0000" }, "message": "ARM: 7336/1: smp_twd: Don\u0027t register CPUFREQ notifiers if local timers are not initialised\n\nCurrent ARM local timer code registers CPUFREQ notifiers even in case\nthe twd_timer_setup() isn\u0027t called. That seems to be wrong and\nwould eventually lead to kernel crash on the CPU frequency transitions\non the SOCs where the local timer doesn\u0027t exist or broken because of\nhardware BUG. Fix it by testing twd_evt and *__this_cpu_ptr(twd_evt).\n\nThe issue was observed with v3.3-rc3 and building an OMAP2+ kernel\non OMAP3 SOC which doesn\u0027t have TWD.\n\nBelow is the dump for reference :\n\n Unable to handle kernel paging request at virtual address 007e900\n pgd \u003d cdc20000\n [007e9000] *pgd\u003d00000000\n Internal error: Oops: 5 [#1] SMP\n Modules linked in:\n CPU: 0 Not tainted (3.3.0-rc3-pm+debug+initramfs #9)\n PC is at twd_update_frequency+0x34/0x48\n LR is at twd_update_frequency+0x10/0x48\n pc : [\u003cc001382c\u003e] lr : [\u003cc0013808\u003e] psr: 60000093\n sp : ce311dd8 ip : 00000000 fp : 00000000\n r10: 00000000 r9 : 00000001 r8 : ce310000\n r7 : c0440458 r6 : c00137f8 r5 : 00000000 r4 : c0947a74\n r3 : 00000000 r2 : 007e9000 r1 : 00000000 r0 : 00000000\n Flags: nZCv IRQs off FIQs on Mode SVC_32 ISA ARM Segment usr\n Control: 10c5387d Table: 8dc20019 DAC: 00000015\n Process sh (pid: 599, stack limit \u003d 0xce3102f8)\n Stack: (0xce311dd8 to 0xce312000)\n 1dc0: 6000c\n 1de0: 00000001 00000002 00000000 00000000 00000000 00000000 00000\n 1e00: ffffffff c093d8f0 00000000 ce311ebc 00000001 00000001 ce310\n 1e20: c001386c c0437c4c c0e95b60 c0e95ba8 00000001 c0e95bf8 ffff4\n 1e40: 00000000 00000000 c005ef74 ce310000 c0435cf0 ce311ebc 00000\n 1e60: ce352b40 0007a120 c08d5108 c08ba040 c08ba040 c005f030 00000\n 1e80: c08bc554 c032fe2c 0007a120 c08d4b64 ce352b40 c08d8618 ffff8\n 1ea0: c08ba040 c033364c ce311ecc c0433b50 00000002 ffffffea c0330\n 1ec0: 0007a120 0007a120 22222201 00000000 22222222 00000000 ce357\n 1ee0: ce3d6000 cdc2aed8 ce352ba0 c0470164 00000002 c032f47c 00034\n 1f00: c0331cac ce352b40 00000007 c032f6d0 ce352bbc 0003d090 c0930\n 1f20: c093d8bc c03306a4 00000007 ce311f80 00000007 cdc2aec0 ce358\n 1f40: ce8d20c0 00000007 b6fe5000 ce311f80 00000007 ce310000 0000c\n 1f60: c000de74 ce987400 ce8d20c0 b6fe5000 00000000 00000000 0000c\n 1f80: 00000000 00000000 001fbac8 00000000 00000007 001fbac8 00004\n 1fa0: c000df04 c000dd60 00000007 001fbac8 00000001 b6fe5000 00000\n 1fc0: 00000007 001fbac8 00000007 00000004 b6fe5000 00000000 00202\n 1fe0: 00000000 beb565f8 00101ffc 00008e8c 60000010 00000001 00000\n [\u003cc001382c\u003e] (twd_update_frequency+0x34/0x48) from [\u003cc008ac4c\u003e] )\n [\u003cc008ac4c\u003e] (smp_call_function_single+0x17c/0x1c8) from [\u003cc0013)\n [\u003cc0013890\u003e] (twd_cpufreq_transition+0x24/0x30) from [\u003cc0437c4c\u003e)\n [\u003cc0437c4c\u003e] (notifier_call_chain+0x44/0x84) from [\u003cc005efe4\u003e] ()\n [\u003cc005efe4\u003e] (__srcu_notifier_call_chain+0x70/0xa4) from [\u003cc005f)\n [\u003cc005f030\u003e] (srcu_notifier_call_chain+0x18/0x20) from [\u003cc032fe2)\n [\u003cc032fe2c\u003e] (cpufreq_notify_transition+0xc8/0x1b0) from [\u003cc0333)\n [\u003cc033364c\u003e] (omap_target+0x1b4/0x28c) from [\u003cc032f47c\u003e] (__cpuf)\n [\u003cc032f47c\u003e] (__cpufreq_driver_target+0x50/0x64) from [\u003cc0331d24)\n [\u003cc0331d24\u003e] (cpufreq_set+0x78/0x98) from [\u003cc032f6d0\u003e] (store_sc)\n [\u003cc032f6d0\u003e] (store_scaling_setspeed+0x5c/0x74) from [\u003cc03306a4\u003e)\n [\u003cc03306a4\u003e] (store+0x58/0x74) from [\u003cc014d868\u003e] (sysfs_write_fi)\n [\u003cc014d868\u003e] (sysfs_write_file+0x80/0xb4) from [\u003cc00f2c2c\u003e] (vfs)\n [\u003cc00f2c2c\u003e] (vfs_write+0xa8/0x138) from [\u003cc00f2e9c\u003e] (sys_write)\n [\u003cc00f2e9c\u003e] (sys_write+0x40/0x6c) from [\u003cc000dd60\u003e] (ret_fast_s)\n Code: e594300c e792210c e1a01000 e5840004 (e7930002)\n ---[ end trace 5da3b5167c1ecdda ]---\n\nReported-by: Kevin Hilman \u003ckhilman@ti.com\u003e\nAcked-by: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\nTested-by: Kevin Hilman \u003ckhilman@ti.com\u003e\nSigned-off-by: Santosh Shilimkar \u003csantosh.shilimkar@ti.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "27e74da9800289e69ba907777df1e2085231eff7", "tree": "3b01385191da0722f747529908a902b169fae805", "parents": [ "8ebbfb49574818f07a25bc119b1bbdf1c2853a39" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 19:34:10 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 19:34:10 2012 -0800" }, "message": "i387: export \u0027fpu_owner_task\u0027 per-cpu variable\n\n(And define it properly for x86-32, which had its \u0027current_task\u0027\ndeclaration in separate from x86-64)\n\nBitten by my dislike for modules on the machines I use, and the fact\nthat apparently nobody else actually wanted to test the patches I sent\nout.\n\nSnif. Nobody else cares.\n\nAnyway, we probably should uninline the \u0027kernel_fpu_begin()\u0027 function\nthat is what modules actually use and that references this, but this is\nthe minimal fix for now.\n\nReported-by: Josh Boyer \u003cjwboyer@gmail.com\u003e\nReported-and-tested-by: Jongman Heo \u003cjongman.heo@samsung.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8ebbfb49574818f07a25bc119b1bbdf1c2853a39", "tree": "e7b1ea6c6a3ea522654c685f2f0c9f41466fdcf9", "parents": [ "39e255dab5a993cbebb35598015da1d4a0eb3727", "847c9db5cb50841589b8ebd3da0769b1b02fb3b2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 16:13:58 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 16:13:58 2012 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nAssorted fixes, sat in -next for a week or so...\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:\n ocfs2: deal with wraparounds of i_nlink in ocfs2_rename()\n vfs: fix compat_sys_stat() handling of overflows in st_nlink\n quota: Fix deadlock with suspend and quotas\n vfs: Provide function to get superblock and wait for it to thaw\n vfs: fix panic in __d_lookup() with high dentry hashtable counts\n autofs4 - fix lockdep splat in autofs\n vfs: fix d_inode_lookup() dentry ref leak\n" }, { "commit": "39e255dab5a993cbebb35598015da1d4a0eb3727", "tree": "3994bee56175ff08d09a6b2de5783456252833d6", "parents": [ "1a4edd9072d3826f1b1234a9b3cc69fcfdebdbfa", "cf1eb40f8f5ea12c9e569e7282161fc7f194fd62" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 16:13:39 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 16:13:39 2012 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:\n [S390] correct ktime to tod clock comparator conversion\n [S390] 3215 deadlock with tty_wakeup\n [S390] incorrect PageTables counter for kvm page tables\n [S390] idle: avoid RCU usage in extended quiescent state\n" }, { "commit": "1a4edd9072d3826f1b1234a9b3cc69fcfdebdbfa", "tree": "f57a889e4b66b3f5069321ce845655637000a477", "parents": [ "7e16838d94b566a17b65231073d179bc04d590c8", "59cca653a601372e9b4a430d867377a3e4a36d76" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 16:13:23 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 16:13:23 2012 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:\n digsig: changed type of the timestamp\n" }, { "commit": "7e16838d94b566a17b65231073d179bc04d590c8", "tree": "356ae3999d89d2419fd4b85b062a24820f4a4d82", "parents": [ "80ab6f1e8c981b1b6604b2f22e36c917526235cd" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 19 13:27:00 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 10:58:54 2012 -0800" }, "message": "i387: support lazy restore of FPU state\n\nThis makes us recognize when we try to restore FPU state that matches\nwhat we already have in the FPU on this CPU, and avoids the restore\nentirely if so.\n\nTo do this, we add two new data fields:\n\n - a percpu \u0027fpu_owner_task\u0027 variable that gets written any time we\n update the \"has_fpu\" field, and thus acts as a kind of back-pointer\n to the task that owns the CPU. The exception is when we save the FPU\n state as part of a context switch - if the save can keep the FPU\n state around, we leave the \u0027fpu_owner_task\u0027 variable pointing at the\n task whose FP state still remains on the CPU.\n\n - a per-thread \u0027last_cpu\u0027 field, that indicates which CPU that thread\n used its FPU on last. We update this on every context switch\n (writing an invalid CPU number if the last context switch didn\u0027t\n leave the FPU in a lazily usable state), so we know that *that*\n thread has done nothing else with the FPU since.\n\nThese two fields together can be used when next switching back to the\ntask to see if the CPU still matches: if \u0027fpu_owner_task\u0027 matches the\ntask we are switching to, we know that no other task (or kernel FPU\nusage) touched the FPU on this CPU in the meantime, and if the current\nCPU number matches the \u0027last_cpu\u0027 field, we know that this thread did no\nother FP work on any other CPU, so the FPU state on the CPU must match\nwhat was saved on last context switch.\n\nIn that case, we can avoid the \u0027f[x]rstor\u0027 entirely, and just clear the\nCR0.TS bit.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "80ab6f1e8c981b1b6604b2f22e36c917526235cd", "tree": "dfd8c2e909d614bc230aa87c0ea5742cf8510c57", "parents": [ "cea20ca3f3181fc36788a15bc65d1062b96a0a6c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 19 11:48:44 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 10:58:28 2012 -0800" }, "message": "i387: use \u0027restore_fpu_checking()\u0027 directly in task switching code\n\nThis inlines what is usually just a couple of instructions, but more\nimportantly it also fixes the theoretical error case (can that FPU\nrestore really ever fail? Maybe we should remove the checking).\n\nWe can\u0027t start sending signals from within the scheduler, we\u0027re much too\ndeep in the kernel and are holding the runqueue lock etc. So don\u0027t\nbother even trying.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "cea20ca3f3181fc36788a15bc65d1062b96a0a6c", "tree": "86c646c0d228d1fe265536b81d86af406ff92c8e", "parents": [ "b01543dfe67bb1d191998e90d20534dc354de059" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 10:24:09 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 20 10:24:09 2012 -0800" }, "message": "i387: fix up some fpu_counter confusion\n\nThis makes sure we clear the FPU usage counter for newly created tasks,\njust so that we start off in a known state (for example, don\u0027t try to\npreload the FPU state on the first task switch etc).\n\nIt also fixes a thinko in when we increment the fpu_counter at task\nswitch time, introduced by commit 34ddc81a230b (\"i387: re-introduce FPU\nstate preloading at context switch time\"). We should increment the\n*new* task fpu_counter, not the old task, and only if we decide to use\nthat state (whether lazily or preloaded).\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "59cca653a601372e9b4a430d867377a3e4a36d76", "tree": "865c5d380ee4f6ca8ee146fc7f00273f357a109b", "parents": [ "b01543dfe67bb1d191998e90d20534dc354de059" ], "author": { "name": "Dmitry Kasatkin", "email": "dmitry.kasatkin@intel.com", "time": "Thu Feb 02 10:46:49 2012 +0200" }, "committer": { "name": "James Morris", "email": "jmorris@namei.org", "time": "Mon Feb 20 19:46:36 2012 +1100" }, "message": "digsig: changed type of the timestamp\n\ntime_t was used in the signature and key packet headers,\nwhich is typedef of long and is different on 32 and 64 bit architectures.\nSignature and key format should be independent of architecture.\nSimilar to GPG, I have changed the type to uint32_t.\n\nSigned-off-by: Dmitry Kasatkin \u003cdmitry.kasatkin@intel.com\u003e\nSigned-off-by: James Morris \u003cjmorris@namei.org\u003e\n" }, { "commit": "b01543dfe67bb1d191998e90d20534dc354de059", "tree": "15103cae057915a9f20bd2ba91f06ae754d4e9ff", "parents": [ "be2874cb4e625716fa9758ad122084da7764cad0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:53:33 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:53:33 2012 -0800" }, "message": "Linux 3.3-rc4\n" }, { "commit": "be2874cb4e625716fa9758ad122084da7764cad0", "tree": "68966366c698bac9e4fd5dc5d12aa4bc3aad98fc", "parents": [ "584216b79ca684431ac6f961ae5cf0bb1d31024f", "a5368e770c9ec58b9d18378844c149df8513e7b8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:40:00 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:40:00 2012 -0800" }, "message": "Merge tag \u0027fixes-3.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc\n\nThese are the bug fixes that have accumulated since 3.3-rc3 in arm-soc.\nThe majority of them are regression fixes for stuff that broke during\nthe merge 3.3 window.\n\nThe notable ones are:\n\n* The at91 ata drivers both broke because of an earlier cleanup patch that\n some other patches were based on. Jean-Christophe decided to remove\n the legacy at91_ide driver and fix the new-style at91-pata driver while\n keeping the cleanup patch. I almost rejected the patches for being too\n late and too big but in the end decided to accept them because they\n fix a regression.\n\n* A patch fixing build breakage from the sysdev-to-device conversion\n colliding with other changes touches a number of mach-s3c files.\n\n* b0654037 \"ARM: orion: Fix Orion5x GPIO regression from MPP cleanup\"\n is a mechanical change that unfortunately touches a lot of lines\n that should up in the diffstat.\n\n* tag \u0027fixes-3.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (28 commits)\n ARM: at91: drop ide driver in favor of the pata one\n pata/at91: use newly introduced SMC accessors\n ARM: at91: add accessor to manage SMC\n ARM: at91:rtc/rtc-at91sam9: ioremap register bank\n ARM: at91: USB AT91 gadget registration for module\n ep93xx: fix build of vision_ep93xx.c\n ARM: OMAP2xxx: PM: fix OMAP2xxx-specific UART idle bug in v3.3\n ARM: orion: Fix USB phy for orion5x.\n ARM: orion: Fix Orion5x GPIO regression from MPP cleanup\n ARM: EXYNOS: Add cpu-offset property in gic device tree node\n ARM: EXYNOS: Bring exynos4-dt up to date\n ARM: OMAP3: cm-t35: fix section mismatch warning\n ARM: OMAP2: Fix the OMAP2 only build break seen with 2011+ ARM tool-chains\n ARM: tegra: paz00: fix wrong UART port on mini-pcie plug\n ARM: tegra: paz00: fix wrong SD1 power gpio\n i2c: tegra: Add devexit_p() for remove\n ARM: EXYNOS: Correct M-5MOLS sensor clock frequency on Universal C210 board\n ARM: EXYNOS: Correct framebuffer window size on Nuri board\n ARM: SAMSUNG: Fix missing api-change from subsys_interface change\n ARM: EXYNOS: Fix \"warning: initialization from incompatible pointer type\"\n ...\n" }, { "commit": "584216b79ca684431ac6f961ae5cf0bb1d31024f", "tree": "263853bef1aacfd7626ef7471faccbe0ffb9626d", "parents": [ "bff98bfcdb2124e4e640fa6d7c9c7c3d21362760", "6c23e4132258be41218584cfb37a43d9346cdd64" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:38:12 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:38:12 2012 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net\n\n1) VETH_INFO_PEER netlink attribute needs to have it\u0027s size validated,\n from Thomas Graf.\n\n2) \u0027poll\u0027 module option of bnx2x driver crashes the machine, just remove\n it. From Michal Schmidt.\n\n3) ks8851_mll driver reads the irq number from two places, but only\n initializes one of them, oops. Use only one location and fix this\n problem, from Jan Weitzel.\n\n4) Fix buffer overrun and unicast sterring bugs in mellanox mlx4 driver,\n from Eugenia Emantayev.\n\n5) Swapped kcalloc() args in RxRPC and mlx4, from Axel Lin.\n\n6) PHY MDIO device name regression fixes from Florian Fainelli.\n\n7) If the wake event IRQ line is different from the netdevice one, we\n have to properly route it to the stmmac interrupt handler. From\n Francesco Virlinzi.\n\n8) Fix rwlock lock initialization ordering bug in mac80211, from\n Mohammed Shafi Shajakhan.\n\n9) TCP lost_cnt can get out of sync, and in fact go negative, in certain\n circumstances. Fix the way we specify what sequence range to operate\n on in tcp_sacktag_one() to fix this bug. From Neal Cardwell.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (27 commits)\n net/ethernet: ks8851_mll fix irq handling\n veth: Enforce minimum size of VETH_INFO_PEER\n stmmac: update the driver version to Feb 2012 (v2)\n stmmac: move hw init in the probe (v2)\n stmmac: request_irq when use an ext wake irq line (v2)\n stmmac: do not discard frame on dribbling bit assert\n ipheth: Add iPhone 4S\n mlx4: add unicast steering entries to resource_tracker\n mlx4: fix QP tree trashing\n mlx4: fix buffer overrun\n 3c59x: shorten timer period for slave devices\n netpoll: netpoll_poll_dev() should access dev-\u003eflags\n RxRPC: Fix kcalloc parameters swapped\n bnx2x: remove the \u0027poll\u0027 module option\n tcp: fix tcp_shifted_skb() adjustment of lost_cnt_hint for FACK\n ks8851: Fix NOHZ local_softirq_pending 08 warning\n bnx2x: fix bnx2x_storm_stats_update() on big endian\n ixp4xx-eth: fix PHY name to match MDIO bus name\n octeon: fix PHY name to match MDIO bus name\n fec: fix PHY name to match fixed MDIO bus name\n ...\n" }, { "commit": "bff98bfcdb2124e4e640fa6d7c9c7c3d21362760", "tree": "6d94d9e8ab057fb85cd634e15a1ddd15df554d30", "parents": [ "4686066689482214781573fcb42300f713d0210c", "61cddc57dc14a5dffa0921d9a24fd68edbb374ac" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:37:25 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:37:25 2012 -0800" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap\n\nFixes a bootstrapping issue for some registers when a less commonly used\nmethod for register cache initialisation is used. Only affects a fairly\nsmall proportion of users that both don\u0027t use explicit register defaults\nand do use the cache.\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap:\n regmap: Fix cache defaults initialization from raw cache defaults\n" }, { "commit": "4686066689482214781573fcb42300f713d0210c", "tree": "dd84c7533f4931d9eb6cb5ecea2f054851fc9fce", "parents": [ "7857b996c28c70e7a2c25930602f5fae4fe2b091", "465c9343c5b746ec2325a220fa3e50cc647d2db7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:28:56 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:28:56 2012 -0800" }, "message": "Merge tag \u0027ecryptfs-3.3-rc4-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tyhicks/ecryptfs\n\nFixes maximum filename length and filesystem type reporting in statfs() calls\nand also fixes stale inode mode bits on eCryptfs inodes after a POSIX ACL was\nset on the lower filesystem\u0027s inode.\n\n* tag \u0027ecryptfs-3.3-rc4-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tyhicks/ecryptfs:\n ecryptfs: remove the second argument of k[un]map_atomic()\n eCryptfs: Copy up lower inode attrs after setting lower xattr\n eCryptfs: Improve statfs reporting\n" }, { "commit": "7857b996c28c70e7a2c25930602f5fae4fe2b091", "tree": "3011fd3dd7be12b4811ee8ae73b154a8dc202944", "parents": [ "06ca7c4376b779e6cb11527abc4f609ddca35b58", "8dc6ae4d448758a30cf5fa822d6fe6f4e15a04c6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:27:40 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:27:40 2012 -0800" }, "message": "Merge tag \u0027pinctrl\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl\n\npinctrl fixes for v3.3\n\n* tag \u0027pinctrl-for-torvalds-20120216\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:\n pinctrl: restore pin naming\n" }, { "commit": "06ca7c4376b779e6cb11527abc4f609ddca35b58", "tree": "3762945d93d82c1489d1ef254941d2af0c5fdc57", "parents": [ "7bcd5b4671c3aa9185a7e6f76ee12e392070ddea", "9a45a9407c69d068500923480884661e2b9cc421" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:26:37 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:26:37 2012 -0800" }, "message": "Merge branch \u0027merge\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc\n\nHere are a few more fixes for powerpc. Some are regressions, the rest\nis simple/obvious/nasty enough that I deemed it good to go now.\n\nHere\u0027s also step one of deprecating legacy iSeries support: we are\nremoving it from the main defconfig.\n\nNobody seems to be using it anymore and the code is nasty to maintain,\n(involves horrible hacks in various low level areas of the kernel) so we\nplan to actually rip it out at some point. For now let\u0027s just avoid\nbuilding it by default. Stephen will proceed to do the actual removal\nlater (probably 3.4 or 3.5).\n\n* \u0027merge\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:\n powerpc/perf: power_pmu_start restores incorrect values, breaking frequency events\n powerpc/adb: Use set_current_state()\n powerpc: Disable interrupts early in Program Check\n powerpc: Remove legacy iSeries from ppc64_defconfig\n powerpc/fsl/pci: Fix PCIe fixup regression\n powerpc: Fix kernel log of oops/panic instruction dump\n" }, { "commit": "7bcd5b4671c3aa9185a7e6f76ee12e392070ddea", "tree": "db80f1cc57cdec18e46bf8e37691b02f9a2ce35e", "parents": [ "58e44bafbb601b5abc250cdbb032b1f4022ef30a", "3682a3946d2b0bad621db871e3bead83e523a238" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:26:11 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:26:11 2012 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci\n\nOne regression fix for SR-IOV on PPC and a couple of misc fixes from\nYinghai.\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci:\n PCI: Fix pci cardbus removal\n PCI: set pci sriov page size before reading SRIOV BAR\n PCI: workaround hard-wired bus number V2\n" }, { "commit": "58e44bafbb601b5abc250cdbb032b1f4022ef30a", "tree": "e0b618339c64cf1a372e299a9f262dd7670c0b6e", "parents": [ "a18d3afefa0104419b5e069af5922bb57a302426", "b7f5b7dec3d539a84734f2bcb7e53fbb1532a40b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:25:39 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:25:39 2012 -0800" }, "message": "Merge branch \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux\n\n3 radeon fixes, I have some exynos fixes to push later but I\u0027ll queue\nthem separately once I\u0027ve looked them over a bit.\n\n* \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux:\n drm/radeon/kms: fix MSI re-arm on rv370+\n drm/radeon/kms/atom: bios scratch reg handling updates\n drm/radeon/kms: drop lock in return path of radeon_fence_count_emitted.\n" }, { "commit": "a18d3afefa0104419b5e069af5922bb57a302426", "tree": "4bcc6010e2eb46d5d40454c4c9276f1ec6c7fd73", "parents": [ "34ddc81a230b15c0e345b6b253049db731499f7e", "f2ea0f5f04c97b48c88edccba52b0682fbe45087" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:24:05 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 15:24:05 2012 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:\n crypto: sha512 - use standard ror64()\n" }, { "commit": "34ddc81a230b15c0e345b6b253049db731499f7e", "tree": "0c3afd68071ec1a8a1d8724ef9a42ef845ecf402", "parents": [ "f94edacf998516ac9d849f7bc6949a703977a7f3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 12:56:35 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 14:03:48 2012 -0800" }, "message": "i387: re-introduce FPU state preloading at context switch time\n\nAfter all the FPU state cleanups and finally finding the problem that\ncaused all our FPU save/restore problems, this re-introduces the\npreloading of FPU state that was removed in commit b3b0870ef3ff (\"i387:\ndo not preload FPU state at task switch time\").\n\nHowever, instead of simply reverting the removal, this reimplements\npreloading with several fixes, most notably\n\n - properly abstracted as a true FPU state switch, rather than as\n open-coded save and restore with various hacks.\n\n In particular, implementing it as a proper FPU state switch allows us\n to optimize the CR0.TS flag accesses: there is no reason to set the\n TS bit only to then almost immediately clear it again. CR0 accesses\n are quite slow and expensive, don\u0027t flip the bit back and forth for\n no good reason.\n\n - Make sure that the same model works for both x86-32 and x86-64, so\n that there are no gratuitous differences between the two due to the\n way they save and restore segment state differently due to\n architectural differences that really don\u0027t matter to the FPU state.\n\n - Avoid exposing the \"preload\" state to the context switch routines,\n and in particular allow the concept of lazy state restore: if nothing\n else has used the FPU in the meantime, and the process is still on\n the same CPU, we can avoid restoring state from memory entirely, just\n re-expose the state that is still in the FPU unit.\n\n That optimized lazy restore isn\u0027t actually implemented here, but the\n infrastructure is set up for it. Of course, older CPU\u0027s that use\n \u0027fnsave\u0027 to save the state cannot take advantage of this, since the\n state saving also trashes the state.\n\nIn other words, there is now an actual _design_ to the FPU state saving,\nrather than just random historical baggage. Hopefully it\u0027s easier to\nfollow as a result.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f94edacf998516ac9d849f7bc6949a703977a7f3", "tree": "4a5c5487d86f3b3873b8ca1ba1c0b00b832870ad", "parents": [ "4903062b5485f0e2c286a23b44c9b59d9b017d53" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 17 21:48:54 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 18 10:19:41 2012 -0800" }, "message": "i387: move TS_USEDFPU flag from thread_info to task_struct\n\nThis moves the bit that indicates whether a thread has ownership of the\nFPU from the TS_USEDFPU bit in thread_info-\u003estatus to a word of its own\n(called \u0027has_fpu\u0027) in task_struct-\u003ethread.has_fpu.\n\nThis fixes two independent bugs at the same time:\n\n - changing \u0027thread_info-\u003estatus\u0027 from the scheduler causes nasty\n problems for the other users of that variable, since it is defined to\n be thread-synchronous (that\u0027s what the \"TS_\" part of the naming was\n supposed to indicate).\n\n So perfectly valid code could (and did) do\n\n\tti-\u003estatus |\u003d TS_RESTORE_SIGMASK;\n\n and the compiler was free to do that as separate load, or and store\n instructions. Which can cause problems with preemption, since a task\n switch could happen in between, and change the TS_USEDFPU bit. The\n change to TS_USEDFPU would be overwritten by the final store.\n\n In practice, this seldom happened, though, because the \u0027status\u0027 field\n was seldom used more than once, so gcc would generally tend to\n generate code that used a read-modify-write instruction and thus\n happened to avoid this problem - RMW instructions are naturally low\n fat and preemption-safe.\n\n - On x86-32, the current_thread_info() pointer would, during interrupts\n and softirqs, point to a *copy* of the real thread_info, because\n x86-32 uses %esp to calculate the thread_info address, and thus the\n separate irq (and softirq) stacks would cause these kinds of odd\n thread_info copy aliases.\n\n This is normally not a problem, since interrupts aren\u0027t supposed to\n look at thread information anyway (what thread is running at\n interrupt time really isn\u0027t very well-defined), but it confused the\n heck out of irq_fpu_usable() and the code that tried to squirrel\n away the FPU state.\n\n (It also caused untold confusion for us poor kernel developers).\n\nIt also turns out that using \u0027task_struct\u0027 is actually much more natural\nfor most of the call sites that care about the FPU state, since they\ntend to work with the task struct for other reasons anyway (ie\nscheduling). And the FPU data that we are going to save/restore is\nfound there too.\n\nThanks to Arjan Van De Ven \u003carjan@linux.intel.com\u003e for pointing us to\nthe %esp issue.\n\nCc: Arjan van de Ven \u003carjan@linux.intel.com\u003e\nReported-and-tested-by: Raphael Prevost \u003craphael@buro.asia\u003e\nAcked-and-tested-by: Suresh Siddha \u003csuresh.b.siddha@intel.com\u003e\nTested-by: Peter Anvin \u003chpa@zytor.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "abe9a6d57b4544ac208401f9c0a4262814db2be4", "tree": "7fb4e9f48bcc6f9269ff59be77c7d9ad9c15ed2d", "parents": [ "f86f36a6ae625eda87a13e1ea102a908e08f491b" ], "author": { "name": "Weston Andros Adamson", "email": "dros@netapp.com", "time": "Thu Feb 16 11:17:05 2012 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Feb 17 17:34:03 2012 -0500" }, "message": "NFSv4: fix server_scope memory leak\n\nserver_scope would never be freed if nfs4_check_cl_exchange_flags() returned\nnon-zero\n\nSigned-off-by: Weston Andros Adamson \u003cdros@netapp.com\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "f86f36a6ae625eda87a13e1ea102a908e08f491b", "tree": "4facbd3378390f4a571bbc169e197a3373f39159", "parents": [ "b9f9a03150969e4bd9967c20bce67c4de769058f" ], "author": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Tue Feb 14 20:33:19 2012 -0500" }, "committer": { "name": "Trond Myklebust", "email": "Trond.Myklebust@netapp.com", "time": "Fri Feb 17 17:33:39 2012 -0500" }, "message": "NFSv4.1: Fix a NFSv4.1 session initialisation regression\n\nCommit aacd553 (NFSv4.1: cleanup init and reset of session slot tables)\nintroduces a regression in the session initialisation code. New tables\nnow find their sequence ids initialised to 0, rather than the mandated\nvalue of 1 (see RFC5661).\n\nFix the problem by merging nfs4_reset_slot_table() and nfs4_init_slot_table().\nSince the tbl-\u003emax_slots is initialised to 0, the test in\nnfs4_reset_slot_table for max_reqs !\u003d tbl-\u003emax_slots will automatically\npass for an empty table.\n\nReported-by: Vitaliy Gusev \u003cgusev.vitaliy@nexenta.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n" }, { "commit": "cf1eb40f8f5ea12c9e569e7282161fc7f194fd62", "tree": "768a78bd496b7a385bd2b05d5dde4563c84728cd", "parents": [ "656d9125376006cf696b0836f1c6723a892629ca" ], "author": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Fri Feb 17 10:29:23 2012 +0100" }, "committer": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Fri Feb 17 10:29:33 2012 +0100" }, "message": "[S390] correct ktime to tod clock comparator conversion\n\nThe conversion of the ktime to a value suitable for the clock comparator\ndoes not take changes to wall_to_monotonic into account. In fact the\nconversion just needs the boot clock (sched_clock_base_cc) and the\ntotal_sleep_time.\n\nThis is applicable to 3.2+ kernels.\n\nCC: stable@vger.kernel.org\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\n" }, { "commit": "656d9125376006cf696b0836f1c6723a892629ca", "tree": "0e5c7b817c01c408dc6edf28602c31a8b4fef299", "parents": [ "2320c5793790fcda80e6dcc088dbda86040235e5" ], "author": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Fri Feb 17 10:29:22 2012 +0100" }, "committer": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Fri Feb 17 10:29:33 2012 +0100" }, "message": "[S390] 3215 deadlock with tty_wakeup\n\nThe 3215 driver calls tty_wakeup from irq context while holding the\ndevice spinlock. If printk is called by any function on the callchain\nstarting from tty_wakeup the system deadlocks on the device spinlock.\nUsing a tasklet to call tty_wakup solves the problem.\n\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\n" }, { "commit": "2320c5793790fcda80e6dcc088dbda86040235e5", "tree": "647d98c36f114a48ee02a738c720ed2f8355a9a9", "parents": [ "f3612304ee04a1a36ded7604771ea56d818158cb" ], "author": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Fri Feb 17 10:29:21 2012 +0100" }, "committer": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Fri Feb 17 10:29:33 2012 +0100" }, "message": "[S390] incorrect PageTables counter for kvm page tables\n\nThe page_table_free_pgste function is used for kvm processes to free page\ntables that have the pgste extension. It calls pgtable_page_ctor instead of\npgtable_page_dtor which increases NR_PAGETABLE instead of decreasing it.\n\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\n" }, { "commit": "f3612304ee04a1a36ded7604771ea56d818158cb", "tree": "3be2e76a2a7929890b02bad3e6ed189419def3bc", "parents": [ "4903062b5485f0e2c286a23b44c9b59d9b017d53" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Fri Feb 17 10:29:20 2012 +0100" }, "committer": { "name": "Martin Schwidefsky", "email": "schwidefsky@de.ibm.com", "time": "Fri Feb 17 10:29:32 2012 +0100" }, "message": "[S390] idle: avoid RCU usage in extended quiescent state\n\nAvoid calling wake_up() from our NMI \"bottom halve\" from RCU extended\nquiescent state in idle. wake_up() has RCU read-side critical sections\nbut this will be completely ignored by RCU if the cpu is in extended\nquiescent state.\nWhich means that whatever object is being accessed from within the\nread-side critical section can be freed concurrently from a different\ncpu.\nSo make sure we leave extended quiescent state before calling wake_up().\n\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\n" }, { "commit": "4903062b5485f0e2c286a23b44c9b59d9b017d53", "tree": "c521dd28c5aa409dcd76ca8a522886fa3c272a31", "parents": [ "b3b0870ef3ffed72b92415423da864f440f57ad6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 19:11:15 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 19:11:15 2012 -0800" }, "message": "i387: move AMD K7/K8 fpu fxsave/fxrstor workaround from save to restore\n\nThe AMD K7/K8 CPUs don\u0027t save/restore FDP/FIP/FOP unless an exception is\npending. In order to not leak FIP state from one process to another, we\nneed to do a floating point load after the fxsave of the old process,\nand before the fxrstor of the new FPU state. That resets the state to\nthe (uninteresting) kernel load, rather than some potentially sensitive\nuser information.\n\nWe used to do this directly after the FPU state save, but that is\nactually very inconvenient, since it\n\n (a) corrupts what is potentially perfectly good FPU state that we might\n want to lazy avoid restoring later and\n\n (b) on x86-64 it resulted in a very annoying ordering constraint, where\n \"__unlazy_fpu()\" in the task switch needs to be delayed until after\n the DS segment has been reloaded just to get the new DS value.\n\nCoupling it to the fxrstor instead of the fxsave automatically avoids\nboth of these issues, and also ensures that we only do it when actually\nnecessary (the FP state after a save may never actually get used). It\u0027s\nsimply a much more natural place for the leaked state cleanup.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b3b0870ef3ffed72b92415423da864f440f57ad6", "tree": "b3e128019581669d44e6634d3b1bfb169c73598d", "parents": [ "6d59d7a9f5b723a7ac1925c136e93ec83c0c3043" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 15:45:23 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 15:45:23 2012 -0800" }, "message": "i387: do not preload FPU state at task switch time\n\nYes, taking the trap to re-load the FPU/MMX state is expensive, but so\nis spending several days looking for a bug in the state save/restore\ncode. And the preload code has some rather subtle interactions with\nboth paravirtualization support and segment state restore, so it\u0027s not\nnearly as simple as it should be.\n\nAlso, now that we no longer necessarily depend on a single bit (ie\nTS_USEDFPU) for keeping track of the state of the FPU, we migth be able\nto do better. If we are really switching between two processes that\nkeep touching the FP state, save/restore is inevitable, but in the case\nof having one process that does most of the FPU usage, we may actually\nbe able to do much better than the preloading.\n\nIn particular, we may be able to keep track of which CPU the process ran\non last, and also per CPU keep track of which process\u0027 FP state that CPU\nhas. For modern CPU\u0027s that don\u0027t destroy the FPU contents on save time,\nthat would allow us to do a lazy restore by just re-enabling the\nexisting FPU state - with no restore cost at all!\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "465c9343c5b746ec2325a220fa3e50cc647d2db7", "tree": "fbff1765ebac62a5910ba34c15deeb8729fce1be", "parents": [ "545d680938be1e86a6c5250701ce9abaf360c495" ], "author": { "name": "Cong Wang", "email": "amwang@redhat.com", "time": "Fri Feb 10 13:39:50 2012 +0800" }, "committer": { "name": "Tyler Hicks", "email": "tyhicks@canonical.com", "time": "Thu Feb 16 16:06:27 2012 -0600" }, "message": "ecryptfs: remove the second argument of k[un]map_atomic()\n\nSigned-off-by: Cong Wang \u003camwang@redhat.com\u003e\nSigned-off-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\n" }, { "commit": "545d680938be1e86a6c5250701ce9abaf360c495", "tree": "c2cb3c9cfbee7c041d137296b72f311179a41f0c", "parents": [ "4a26620df451ad46151ad21d711ed43e963c004e" ], "author": { "name": "Tyler Hicks", "email": "tyhicks@canonical.com", "time": "Tue Feb 07 17:55:40 2012 -0600" }, "committer": { "name": "Tyler Hicks", "email": "tyhicks@canonical.com", "time": "Thu Feb 16 16:06:27 2012 -0600" }, "message": "eCryptfs: Copy up lower inode attrs after setting lower xattr\n\nAfter passing through a -\u003esetxattr() call, eCryptfs needs to copy the\ninode attributes from the lower inode to the eCryptfs inode, as they\nmay have changed in the lower filesystem\u0027s -\u003esetxattr() path.\n\nOne example is if an extended attribute containing a POSIX Access\nControl List is being set. The new ACL may cause the lower filesystem to\nmodify the mode of the lower inode and the eCryptfs inode would need to\nbe updated to reflect the new mode.\n\nhttps://launchpad.net/bugs/926292\n\nSigned-off-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\nReported-by: Sebastien Bacher \u003cseb128@ubuntu.com\u003e\nCc: John Johansen \u003cjohn.johansen@canonical.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\n" }, { "commit": "4a26620df451ad46151ad21d711ed43e963c004e", "tree": "bf035ea6a656b3e19fe93bf37991632cad96d971", "parents": [ "c38e23456278e967f094b08247ffc3711b1029b2" ], "author": { "name": "Tyler Hicks", "email": "tyhicks@canonical.com", "time": "Sat Nov 05 13:45:08 2011 -0400" }, "committer": { "name": "Tyler Hicks", "email": "tyhicks@canonical.com", "time": "Thu Feb 16 16:06:21 2012 -0600" }, "message": "eCryptfs: Improve statfs reporting\n\nstatfs() calls on eCryptfs files returned the wrong filesystem type and,\nwhen using filename encryption, the wrong maximum filename length.\n\nIf mount-wide filename encryption is enabled, the cipher block size and\nthe lower filesystem\u0027s max filename length will determine the max\neCryptfs filename length. Pre-tested, known good lengths are used when\nthe lower filesystem\u0027s namelen is 255 and a cipher with 8 or 16 byte\nblock sizes is used. In other, less common cases, we fall back to a safe\nrounded-down estimate when determining the eCryptfs namelen.\n\nhttps://launchpad.net/bugs/885744\n\nSigned-off-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\nReported-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReviewed-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n" }, { "commit": "6d59d7a9f5b723a7ac1925c136e93ec83c0c3043", "tree": "9478d5c1cc3a3f85cc937d8d6ddaad926af7e830", "parents": [ "b6c66418dcad0fcf83cd1d0a39482db37bf4fc41" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 13:33:12 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 13:33:12 2012 -0800" }, "message": "i387: don\u0027t ever touch TS_USEDFPU directly, use helper functions\n\nThis creates three helper functions that do the TS_USEDFPU accesses, and\nmakes everybody that used to do it by hand use those helpers instead.\n\nIn addition, there\u0027s a couple of helper functions for the \"change both\nCR0.TS and TS_USEDFPU at the same time\" case, and the places that do\nthat together have been changed to use those. That means that we have\nfewer random places that open-code this situation.\n\nThe intent is partly to clarify the code without actually changing any\nsemantics yet (since we clearly still have some hard to reproduce bug in\nthis area), but also to make it much easier to use another approach\nentirely to caching the CR0.TS bit for software accesses.\n\nRight now we use a bit in the thread-info \u0027status\u0027 variable (this patch\ndoes not change that), but we might want to make it a full field of its\nown or even make it a per-cpu variable.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b6c66418dcad0fcf83cd1d0a39482db37bf4fc41", "tree": "0a23207a2ab766341fbf60d76b2c10984ec6e5ca", "parents": [ "15d8791cae75dca27bfda8ecfe87dca9379d6bb0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 12:22:48 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 12:22:48 2012 -0800" }, "message": "i387: move TS_USEDFPU clearing out of __save_init_fpu and into callers\n\nTouching TS_USEDFPU without touching CR0.TS is confusing, so don\u0027t do\nit. By moving it into the callers, we always do the TS_USEDFPU next to\nthe CR0.TS accesses in the source code, and it\u0027s much easier to see how\nthe two go hand in hand.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "15d8791cae75dca27bfda8ecfe87dca9379d6bb0", "tree": "e8e2cd53544eda0644ee141b6b0414e34d96f6be", "parents": [ "c38e23456278e967f094b08247ffc3711b1029b2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 09:15:04 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 16 09:15:04 2012 -0800" }, "message": "i387: fix x86-64 preemption-unsafe user stack save/restore\n\nCommit 5b1cbac37798 (\"i387: make irq_fpu_usable() tests more robust\")\nadded a sanity check to the #NM handler to verify that we never cause\nthe \"Device Not Available\" exception in kernel mode.\n\nHowever, that check actually pinpointed a (fundamental) race where we do\ncause that exception as part of the signal stack FPU state save/restore\ncode.\n\nBecause we use the floating point instructions themselves to save and\nrestore state directly from user mode, we cannot do that atomically with\ntesting the TS_USEDFPU bit: the user mode access itself may cause a page\nfault, which causes a task switch, which saves and restores the FP/MMX\nstate from the kernel buffers.\n\nThis kind of \"recursive\" FP state save is fine per se, but it means that\nwhen the signal stack save/restore gets restarted, it will now take the\n\u0027#NM\u0027 exception we originally tried to avoid. With preemption this can\nhappen even without the page fault - but because of the user access, we\ncannot just disable preemption around the save/restore instruction.\n\nThere are various ways to solve this, including using the\n\"enable/disable_page_fault()\" helpers to not allow page faults at all\nduring the sequence, and fall back to copying things by hand without the\nuse of the native FP state save/restore instructions.\n\nHowever, the simplest thing to do is to just allow the #NM from kernel\nspace, but fix the race in setting and clearing CR0.TS that this all\nexposed: the TS bit changes and the TS_USEDFPU bit absolutely have to be\natomic wrt scheduling, so while the actual state save/restore can be\ninterrupted and restarted, the act of actually clearing/setting CR0.TS\nand the TS_USEDFPU bit together must not.\n\nInstead of just adding random \"preempt_disable/enable()\" calls to what\nis already excessively ugly code, this introduces some helper functions\nthat mostly mirror the \"kernel_fpu_begin/end()\" functionality, just for\nthe user state instead.\n\nThose helper functions should probably eventually replace the other\nad-hoc CR0.TS and TS_USEDFPU tests too, but I\u0027ll need to think about it\nsome more: the task switching functionality in particular needs to\nexpose the difference between the \u0027prev\u0027 and \u0027next\u0027 threads, while the\nnew helper functions intentionally were written to only work with\n\u0027current\u0027.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "53ef299f3900bc1deb163b94d4f1cac4f3346152", "tree": "db38a3d7661920e93c8d60a2ad6a498e98b1f64b", "parents": [ "bc41eae2c84694667c1d7747fa28db8e75948ac4" ], "author": { "name": "Inki Dae", "email": "inki.dae@samsung.com", "time": "Wed Feb 15 11:25:22 2012 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 16 09:40:54 2012 +0000" }, "message": "drm/exynos: added postclose to release resource.\n\nSigned-off-by: Inki Dae \u003cinki.dae@samsung.com\u003e\nSigned-off-by: Kyungmin Park \u003ckyungmin.park@samsung.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "bc41eae2c84694667c1d7747fa28db8e75948ac4", "tree": "430d497ad5839cedc8e02bce2f9300eabfc2a342", "parents": [ "c5614ae326c7fdd244d8e0365b8b78d5e3cd5bf4" ], "author": { "name": "Inki Dae", "email": "inki.dae@samsung.com", "time": "Wed Feb 15 11:25:21 2012 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 16 09:40:52 2012 +0000" }, "message": "drm/exynos: removed exynos_drm_fbdev_recreate function.\n\nthis function ins\u0027t needed anymore.\n\nSigned-off-by: Inki Dae \u003cinki.dae@samsung.com\u003e\nSigned-off-by: Kyungmin Park \u003ckyungmin.park@samsung.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "c5614ae326c7fdd244d8e0365b8b78d5e3cd5bf4", "tree": "3e3a21cff4a79e6f701354b48c5b4a650361f5f4", "parents": [ "d081f5660422270856b77bcbbaa312138f556c9e" ], "author": { "name": "Inki Dae", "email": "inki.dae@samsung.com", "time": "Wed Feb 15 11:25:20 2012 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 16 09:40:50 2012 +0000" }, "message": "drm/exynos: fixed page flip issue.\n\nwith vblank_refcount \u003d 1, there was the case that drm_vblank_put\nis called by specific page flip function so this patch fixes the\nissue.\n\nSigned-off-by: Inki Dae \u003cinki.dae@samsung.com\u003e\nSigned-off-by: Kyungmin Park \u003ckyungmin.park@samsung.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "d081f5660422270856b77bcbbaa312138f556c9e", "tree": "1e60867d6c87e9c74369b47a3db1de755f810d2d", "parents": [ "6f811502a473a32a9b892fb35d77b8f91dc7484c" ], "author": { "name": "Inki Dae", "email": "inki.dae@samsung.com", "time": "Wed Feb 15 11:25:19 2012 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 16 09:40:47 2012 +0000" }, "message": "drm/exynos: added possible_clones setup function.\n\nbasically, all crtcs are possible to clone each other.\n\nSigned-off-by: Inki Dae \u003cinki.dae@samsung.com\u003e\nSigned-off-by: Kyungmin Park \u003ckyungmin.park@samsung.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "6f811502a473a32a9b892fb35d77b8f91dc7484c", "tree": "abb55be5f1b70e29179311cd4a8cc4fd7577fb64", "parents": [ "44a0e022b86a8c12ed06c02f52045c8f9f118bb1" ], "author": { "name": "Joonyoung Shim", "email": "jy0922.shim@samsung.com", "time": "Wed Feb 15 11:25:18 2012 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 16 09:40:44 2012 +0000" }, "message": "drm/exynos: removed pageflip_event_list init code when closed.\n\nif one process is terminated by ctrl-c while two processes are\nusing pageflip feature then for last pageflip event,\nuser can\u0027t get poll from kernel side so this patch fixes the problem.\n\nSigned-off-by: Joonyoung Shim \u003cjy0922.shim@samsung.com\u003e\nSigned-off-by: Inki Dae \u003cinki.dae@samsung.com\u003e\nSigned-off-by: Kyoungmin Park \u003ckyungmin.park@samsung.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "44a0e022b86a8c12ed06c02f52045c8f9f118bb1", "tree": "aa4bf658dbc8c7877bd1c826327910b051128812", "parents": [ "1109bf8bcbf455e4cfebce862f9f9fa5a2f386e9" ], "author": { "name": "Joonyoung Shim", "email": "jy0922.shim@samsung.com", "time": "Wed Feb 15 11:25:17 2012 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 16 09:40:43 2012 +0000" }, "message": "drm/exynos: changed priority of mixer layers.\n\nSigned-off-by: Joonyoung Shim \u003cjy0922.shim@samsung.com\u003e\nSigned-off-by: Inki Dae \u003cinki.dae@samsung.com\u003e\nSigned-off-by: Kyungmin Park \u003ckyungmin.park@samsung.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "1109bf8bcbf455e4cfebce862f9f9fa5a2f386e9", "tree": "61edb2982b7479967796613e3eba121715813062", "parents": [ "b7f5b7dec3d539a84734f2bcb7e53fbb1532a40b" ], "author": { "name": "Masanari Iida", "email": "standby24x7@gmail.com", "time": "Tue Feb 14 16:52:41 2012 +0900" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Feb 16 09:37:49 2012 +0000" }, "message": "drm/exynos: Fix typo in exynos_mixer.c\n\nCorrect spelling \"sucessful\" to \"successful\" in\ndrivers/gpu/drm/exynos/exynos_mixer.c\n\nSigned-off-by: Masanari Iida \u003cstandby24x7@gmail.com\u003e\nSigned-off-by: Inki Dae \u003cinki.dae@samsung.com\u003e\nSigned-off-by: Kyungmin Park \u003ckyungmin.park@samsung.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "9a45a9407c69d068500923480884661e2b9cc421", "tree": "8ff43b5fd98ab68352dc4dd0823401a5736afab2", "parents": [ "64f8c13561fbd2f1a8b4a8975b2aee73b561093e" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Feb 15 18:48:22 2012 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Thu Feb 16 16:24:35 2012 +1100" }, "message": "powerpc/perf: power_pmu_start restores incorrect values, breaking frequency events\n\nperf on POWER stopped working after commit e050e3f0a71b (perf: Fix\nbroken interrupt rate throttling). That patch exposed a bug in\nthe POWER perf_events code.\n\nSince the PMCs count upwards and take an exception when the top bit\nis set, we want to write 0x80000000 - left in power_pmu_start. We were\ninstead programming in left which effectively disables the counter\nuntil we eventually hit 0x80000000. This could take seconds or longer.\n\nWith the patch applied I get the expected number of samples:\n\n SAMPLE events: 9948\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nAcked-by: Paul Mackerras \u003cpaulus@samba.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: \u003cstable@kernel.org\u003e\n" }, { "commit": "64f8c13561fbd2f1a8b4a8975b2aee73b561093e", "tree": "96a397f590a064de9987ff1144a293fc2bf3602f", "parents": [ "54321242afe6fcf8b7e589bc21ecf832bc5a206a" ], "author": { "name": "majianpeng", "email": "majianpeng@gmail.com", "time": "Fri Feb 03 14:35:59 2012 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Thu Feb 16 16:15:12 2012 +1100" }, "message": "powerpc/adb: Use set_current_state()\n\nSigned-off-by: majianpeng \u003cmajianpeng@gmail.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "54321242afe6fcf8b7e589bc21ecf832bc5a206a", "tree": "dfc1d3bd73cd1c28421e27bbbbcad60368a69907", "parents": [ "a1a1d1bfc9be2a5ea8cce78ebd0f17c79489f053" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Feb 13 20:42:18 2012 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Thu Feb 16 16:15:10 2012 +1100" }, "message": "powerpc: Disable interrupts early in Program Check\n\nProgram Check exceptions are the result of WARNs, BUGs, some\ntype of breakpoints, kprobe, and other illegal instructions.\n\nWe want interrupts (and thus preemption) to remain disabled\nwhile doing the initial stage of testing the reason and\nbranching off to a debugger or kprobe, so we are still on\nthe original CPU which makes debugging easier in various cases.\n\nThis is how the code was intended, hence the local_irq_enable()\nright in the middle of program_check_exception().\n\nHowever, the assembly exception prologue for that exception was\nincorrectly marked as enabling interrupts, which defeats that\n(and records a redundant enable with lockdep).\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "a1a1d1bfc9be2a5ea8cce78ebd0f17c79489f053", "tree": "78a3d679a281096d34c8437e38547d3fcb0e3b2c", "parents": [ "13635dfdc6aa8d2890e02dc441decfcb4ae63e14" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Tue Feb 14 16:31:09 2012 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Thu Feb 16 16:15:08 2012 +1100" }, "message": "powerpc: Remove legacy iSeries from ppc64_defconfig\n\nSince we are heading towards removing the Legacy iSeries platform, start\nby no longer building it for ppc64_defconfig.\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "13635dfdc6aa8d2890e02dc441decfcb4ae63e14", "tree": "817d08d8b06183bccb28e0c829213f1fceab3b3a", "parents": [ "40c8cefaaf12734327db7199a56e60058d98e7b6" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Feb 14 18:22:20 2012 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Thu Feb 16 16:15:03 2012 +1100" }, "message": "powerpc/fsl/pci: Fix PCIe fixup regression\n\nUpstream changes to the way PHB resources are registered\nbroke the resource fixup for FSL boards.\n\nWe can no longer rely on the resource pointer array for the PHB\u0027s\npci_bus structure, so let\u0027s leave it alone and go straight for\nthe PHB resources instead. This also makes the code generally\nmore readable.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "40c8cefaaf12734327db7199a56e60058d98e7b6", "tree": "639b20e1e2229f3bcc03c5ef92dcfa7ca784cd65", "parents": [ "778a785f02ad846446e91dab49331bd7d853c514" ], "author": { "name": "Ira Snyder", "email": "iws@ovro.caltech.edu", "time": "Fri Jan 06 12:34:07 2012 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Thu Feb 16 16:11:23 2012 +1100" }, "message": "powerpc: Fix kernel log of oops/panic instruction dump\n\nA kernel oops/panic prints an instruction dump showing several\ninstructions before and after the instruction which caused the\noops/panic.\n\nThe code intended that the faulting instruction be enclosed in angle\nbrackets, however a bug caused the faulting instruction to be\ninterpreted by printk() as the message log level.\n\nTo fix this, the KERN_CONT log level is added before the actual text of\nthe printed message.\n\n\u003d\u003d\u003d Before the patch \u003d\u003d\u003d\n\n[ 1081.587266] Instruction dump:\n[ 1081.590236] 7c000110 7c0000f8 5400077c 552907f6 7d290378 992b0003 4e800020 38000001\n[ 1081.598034] 3d20c03a 9009a114 7c0004ac 39200000\n[ 1081.602500] 4e800020 3803ffd0 2b800009\n\n\u003c4\u003e[ 1081.587266] Instruction dump:\n\u003c4\u003e[ 1081.590236] 7c000110 7c0000f8 5400077c 552907f6 7d290378 992b0003 4e800020 38000001\n\u003c4\u003e[ 1081.598034] 3d20c03a 9009a114 7c0004ac 39200000\n\u003c98090000\u003e[ 1081.602500] 4e800020 3803ffd0 2b800009\n\n\u003d\u003d\u003d After the patch \u003d\u003d\u003d\n\n[ 51.385216] Instruction dump:\n[ 51.388186] 7c000110 7c0000f8 5400077c 552907f6 7d290378 992b0003 4e800020 38000001\n[ 51.395986] 3d20c03a 9009a114 7c0004ac 39200000 \u003c98090000\u003e 4e800020 3803ffd0 2b800009\n\n\u003c4\u003e[ 51.385216] Instruction dump:\n\u003c4\u003e[ 51.388186] 7c000110 7c0000f8 5400077c 552907f6 7d290378 992b0003 4e800020 38000001\n\u003c4\u003e[ 51.395986] 3d20c03a 9009a114 7c0004ac 39200000 \u003c98090000\u003e 4e800020 3803ffd0 2b800009\n\nSigned-off-by: Ira W. Snyder \u003ciws@ovro.caltech.edu\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: linuxppc-dev@lists.ozlabs.org\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "f2ea0f5f04c97b48c88edccba52b0682fbe45087", "tree": "53ee15dd4d1a2122d0e571dc8263353acb54f333", "parents": [ "3a92d687c8015860a19213e3c102cad6b722f83c" ], "author": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Sat Jan 14 21:44:49 2012 +0300" }, "committer": { "name": "Herbert Xu", "email": "herbert@gondor.apana.org.au", "time": "Thu Feb 16 12:12:33 2012 +0800" }, "message": "crypto: sha512 - use standard ror64()\n\nUse standard ror64() instead of hand-written.\nThere is no standard ror64, so create it.\n\nThe difference is shift value being \"unsigned int\" instead of uint64_t\n(for which there is no reason). gcc starts to emit native ROR instructions\nwhich it doesn\u0027t do for some reason currently. This should make the code\nfaster.\n\nPatch survives in-tree crypto test and ping flood with hmac(sha512) on.\n\nSigned-off-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nSigned-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n" }, { "commit": "1c8ecf80fdee4e7b23a9e7da7ff9bd59ba2dcf96", "tree": "0cbb7a8b88c100acb3df605fc8908845729765cb", "parents": [ "d71de14ddf423ccc9a2e3f7e37553c99ead20d7c" ], "author": { "name": "Eugeni Dodonov", "email": "eugeni.dodonov@intel.com", "time": "Tue Feb 14 11:44:48 2012 -0200" }, "committer": { "name": "Jesse Barnes", "email": "jbarnes@virtuousgeek.org", "time": "Wed Feb 15 17:43:41 2012 -0800" }, "message": "drm/i915: do not enable RC6p on Sandy Bridge\n\nWith base on latest findings, RC6p seems to be respondible for RC6-related\nissues on Sandy Bridge platform. To work-around those issues, the previous\nsolution was to completely disable RC6 on Sandy Bridge for the past few\nreleases, even if plain RC6 was not giving any issues.\n\nWhat this patch does is preventing RC6p from being enabled on Sandy Bridge\neven if users enable RC6 via a kernel parameter. So it won\u0027t change the\ndefaults in any way, but will ensure that if users do enable RC6 manually\nit won\u0027t break their machines by enabling this extra state.\n\nProper fix for this (enabling specific RC6 states according to the GPU\ngeneration) were proposed for the -next kernel, but we are too late in the\nrelease process now to pick such changes.\n\nAcked-by: Keith Packard \u003ckeithp@keithp.com\u003e\nSigned-off-by: Eugeni Dodonov \u003ceugeni.dodonov@intel.com\u003e\nSigned-off-by: Jesse Barnes \u003cjbarnes@virtuousgeek.org\u003e\n" }, { "commit": "fee6a3c33a8f137f39cd9997b0476411f73576c7", "tree": "45a45c311f2e15629d8b61838b5ad1cd8851d935", "parents": [ "46e33c606af8e0caeeca374103189663d877c0d6" ], "author": { "name": "Olof Johansson", "email": "olof@lixom.net", "time": "Wed Feb 15 21:17:12 2012 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Feb 15 21:10:49 2012 +0000" }, "message": "ARM: 7327/1: need to include asm/system.h in asm/processor.h\n\nFor files that include asm/processor.h but not asm/system.h:\n\narch/arm/mach-msm/include/mach/uncompress.h: In function \u0027putc\u0027:\narch/arm/mach-msm/include/mach/uncompress.h:48:3: error: implicit declaration of function \u0027smp_mb\u0027 [-Werror\u003dimplicit-function-declaration]\n\nIn this case, smp_mb() is from the cpu_relax() call in the msm putc().\n\nIt likely went uncaught when the uncompress.h change went in since the\ndefconfig didn\u0027t enable that code path, but later changes (e76f4750f4:\nARM: debug: arrange Kconfig options more logically) resulted in the\noption being on for msm_defconfig and thus exposed it.\n\nSigned-off-by: Olof Johansson \u003colof@lixom.net\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "46e33c606af8e0caeeca374103189663d877c0d6", "tree": "a1f97d7fd9af85b6f505c166222152192c11c1bb", "parents": [ "4272f98a1ae81709fc5c804c33c044064e419cd9" ], "author": { "name": "Javi Merino", "email": "javi.merino@arm.com", "time": "Wed Feb 15 17:36:39 2012 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Feb 15 21:10:49 2012 +0000" }, "message": "ARM: 7326/2: PL330: fix null pointer dereference in pl330_chan_ctrl()\n\nThis fixes the thrd-\u003ereq_running field being accessed before thrd\nis checked for null. The error was introduced in\n\n abb959f: ARM: 7237/1: PL330: Fix driver freeze\n\nReference: \u003c1326458191-23492-1-git-send-email-mans.rullgard@linaro.org\u003e\n\nCc: stable@kernel.org\nSigned-off-by: Mans Rullgard \u003cmans.rullgard@linaro.org\u003e\nAcked-by: Javi Merino \u003cjavi.merino@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "4272f98a1ae81709fc5c804c33c044064e419cd9", "tree": "0e3af4e730a6f1bbec98f3e4808dd272e8ae8e58", "parents": [ "8e43a905dd574f54c5715d978318290ceafbe275" ], "author": { "name": "Javi Merino", "email": "javi.merino@arm.com", "time": "Wed Nov 16 12:36:39 2011 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Feb 15 21:10:49 2012 +0000" }, "message": "ARM: 7164/3: PL330: Fix the size of the dst_cache_ctrl field\n\ndst_cache_ctrl affects bits 3, 1 and 0 of AWCACHE but it is a 3-bit\nfield in the Channel Control Register (see Table 3-21 of the DMA-330\nTechnical Reference Manual) and should be programmed as such.\n\nReference: \u003c1320244259-10496-3-git-send-email-javi.merino@arm.com\u003e\n\nSigned-off-by: Javi Merino \u003cjavi.merino@arm.com\u003e\nAcked-by: Jassi Brar \u003cjassisinghbrar@gmail.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "8e43a905dd574f54c5715d978318290ceafbe275", "tree": "4d2bc5178293689353675105ee37e1c287848f39", "parents": [ "6e2e340b59d2d4e7b6b7f2c2d02b0d5ca4df6458" ], "author": { "name": "Rabin Vincent", "email": "rabin@rab.in", "time": "Wed Feb 15 16:01:42 2012 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Feb 15 21:09:52 2012 +0000" }, "message": "ARM: 7325/1: fix v7 boot with lockdep enabled\n\nBootup with lockdep enabled has been broken on v7 since b46c0f74657d\n(\"ARM: 7321/1: cache-v7: Disable preemption when reading CCSIDR\").\n\nThis is because v7_setup (which is called very early during boot) calls\nv7_flush_dcache_all, and the save_and_disable_irqs added by that patch\nends up attempting to call into lockdep C code (trace_hardirqs_off())\nwhen we are in no position to execute it (no stack, MMU off).\n\nFix this by using a notrace variant of save_and_disable_irqs. The code\nalready uses the notrace variant of restore_irqs.\n\nReviewed-by: Nicolas Pitre \u003cnico@linaro.org\u003e\nAcked-by: Stephen Boyd \u003csboyd@codeaurora.org\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Rabin Vincent \u003crabin@rab.in\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "6c23e4132258be41218584cfb37a43d9346cdd64", "tree": "2121139424540e89e1f5b2ead40c96db5cc0b12b", "parents": [ "237114384ab22c174ec4641e809f8e6cbcfce774" ], "author": { "name": "Jan Weitzel", "email": "j.weitzel@phytec.de", "time": "Tue Feb 14 21:35:15 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 15:41:10 2012 -0500" }, "message": "net/ethernet: ks8851_mll fix irq handling\n\nThere a two different irq variables ks-\u003eirq and netdev-\u003eirq.\nOnly ks-\u003eirq is set on probe, so disabling irq in ks_start_xmit fails.\n\nThis patches remove ks-\u003eirq from private data and use only netdev-\u003eirq.\n\nTested on a kernel 3.0 based OMAP4430 SMP Board\n\nSigned-off-by: Jan Weitzel \u003cj.weitzel@phytec.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "237114384ab22c174ec4641e809f8e6cbcfce774", "tree": "9571ac28f82724a140367ad754b4946a889f2b70", "parents": [ "78a5249fc9dc0f8c70221751dc79b6d918e5ee9a" ], "author": { "name": "Thomas Graf", "email": "tgraf@suug.ch", "time": "Wed Feb 15 04:09:46 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:59:20 2012 -0500" }, "message": "veth: Enforce minimum size of VETH_INFO_PEER\n\nVETH_INFO_PEER carries struct ifinfomsg plus optional IFLA\nattributes. A minimal size of sizeof(struct ifinfomsg) must be\nenforced or we may risk accessing that struct beyond the limits\nof the netlink message.\n\nSigned-off-by: Thomas Graf \u003ctgraf@suug.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "78a5249fc9dc0f8c70221751dc79b6d918e5ee9a", "tree": "96b29593a2c9042cd42e38e4f40f84b1ddf7afad", "parents": [ "cf3f047b9af49d4ee8abfa31b0ef0e99cbcaf17d" ], "author": { "name": "Giuseppe CAVALLARO", "email": "peppe.cavallaro@st.com", "time": "Wed Feb 15 00:10:40 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:52:10 2012 -0500" }, "message": "stmmac: update the driver version to Feb 2012 (v2)\n\nSigned-off-by: Giuseppe Cavallaro \u003cpeppe.cavallaro@st.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "cf3f047b9af49d4ee8abfa31b0ef0e99cbcaf17d", "tree": "695883ca9df55d62ee83d35fc86a37e47bcb04f6", "parents": [ "7a13f8f5b63652c035147aab5fcba7ee9101f1fb" ], "author": { "name": "Giuseppe CAVALLARO", "email": "peppe.cavallaro@st.com", "time": "Wed Feb 15 00:10:39 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:52:10 2012 -0500" }, "message": "stmmac: move hw init in the probe (v2)\n\nThis patch moves the MAC HW initialization and\nthe HW feature verification from the open to the probe\nfunction as D. Miller suggested.\nSo the patch actually reorganizes and tidies-up some parts of\nthe driver and indeed fixes some problem when tune its HW features.\nThese can be overwritten by looking at the HW cap register at\nrun-time and that generated problems.\n\nSigned-off-by: Giuseppe Cavallaro \u003cpeppe.cavallaro@st.com\u003e\nReviewed-by: Francesco Virlinzi \u003cfrancesco.virlinzi@st.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7a13f8f5b63652c035147aab5fcba7ee9101f1fb", "tree": "7fab8140357475616df961b140a7e4d402231938", "parents": [ "1cc5a735185a963ed9324a65e1f2db02bf52c0d4" ], "author": { "name": "Francesco Virlinzi", "email": "francesco.virlinzi@st.com", "time": "Wed Feb 15 00:10:38 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:52:10 2012 -0500" }, "message": "stmmac: request_irq when use an ext wake irq line (v2)\n\nIn case of we use an external Wake-Up IRQ line\n(priv-\u003ewol_irq !\u003d dev-\u003eirq) we need to invoke the\nrequest_irq.\n\nSigned-off-by: Francesco Virlinzi \u003cfrancesco.virlinzi@st.com\u003e\nSigned-off-by: Giuseppe Cavallaro \u003cpeppe.cavallaro@st.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1cc5a735185a963ed9324a65e1f2db02bf52c0d4", "tree": "5e280cb7f6626080ab4ab8e2be01841fbaf6c4f6", "parents": [ "72ba009b8a159e995e40d3b4e5d7d265acead983" ], "author": { "name": "Giuseppe CAVALLARO", "email": "peppe.cavallaro@st.com", "time": "Wed Feb 15 00:10:37 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:52:10 2012 -0500" }, "message": "stmmac: do not discard frame on dribbling bit assert\n\nIf this bit is set and the CRC error is reset, then the packet is valid.\nOnly report this as stat info.\n\nSigned-off-by: Giuseppe Cavallaro \u003cpeppe.cavallaro@st.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "72ba009b8a159e995e40d3b4e5d7d265acead983", "tree": "bed23845d49de76c6ce0507b28c3b7ab38fbed91", "parents": [ "9f5b6c632eb3b495d5ae701c43e67da930b1f637" ], "author": { "name": "Tim Gardner", "email": "tim.gardner@canonical.com", "time": "Wed Feb 15 07:50:15 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:50:16 2012 -0500" }, "message": "ipheth: Add iPhone 4S\n\nBugLink: http://bugs.launchpad.net/bugs/900802\n\nCc: stable@vger.kernel.org 3.2+\nSigned-off-by: Till Kamppeter \u003ctill.kamppeter@gmail.com\u003e\nSigned-off-by: Tim Gardner \u003ctim.gardner@canonical.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9f5b6c632eb3b495d5ae701c43e67da930b1f637", "tree": "7898b3b5db0f0f8fe08eca1c2ea01372777593ee", "parents": [ "2531188b476c83fe29b1d3498a76e329f4b6f16b" ], "author": { "name": "Eugenia Emantayev", "email": "eugenia@mellanox.co.il", "time": "Wed Feb 15 06:23:16 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:50:16 2012 -0500" }, "message": "mlx4: add unicast steering entries to resource_tracker\n\nAdd unicast steering entries to resource tracker.\nDo qp_detach also for these entries when VF doesn\u0027t shut down gracefully.\nOtherwise there is leakage of these resources, since they are not tracked.\n\nSigned-off-by: Eugenia Emantayev \u003ceugenia@mellanox.co.il\u003e\nReviewed-by: Yevgeny Petrilin \u003cyevgenyp@mellanox.co.il\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2531188b476c83fe29b1d3498a76e329f4b6f16b", "tree": "304f540bab2412ce662ab35b3f7d9d08ed2e42f6", "parents": [ "75c6062cb797afe624d65c955eb867035622e782" ], "author": { "name": "Eugenia Emantayev", "email": "eugenia@mellanox.co.il", "time": "Wed Feb 15 06:22:57 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:50:16 2012 -0500" }, "message": "mlx4: fix QP tree trashing\n\nWhen adding new unicast steer entry, before moving qp to state ready,\nactually before calling mlx4_RST2INIT_QP_wrapper(), there were added\na lot of entries with local_qpn\u003d0 into radix tree.\nThis fact impacted the get_res() function and proper functioning\nof resource tracker in addition to adding trash entries into radix tree.\n\nSigned-off-by: Eugenia Emantayev \u003ceugenia@mellanox.co.il\u003e\nReviewed-by: Yevgeny Petrilin \u003cyevgenyp@melllanox.co.il\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "75c6062cb797afe624d65c955eb867035622e782", "tree": "756b4b28f3316befae7dcd6d540eb9924a87031d", "parents": [ "33b5d30cd8b200024d55ac5830c9efde1102ef4c" ], "author": { "name": "Eugenia Emantayev", "email": "eugenia@mellanox.co.il", "time": "Wed Feb 15 06:22:49 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 15 14:50:15 2012 -0500" }, "message": "mlx4: fix buffer overrun\n\nWhen passing MLX4_UC_STEER\u003d1 it was translated to value 2\nafter mlx4_QP_ATTACH_wrapper. Therefore in new_steering_entry()\nunicast steer entries were added to index 2 of array of size 2.\nFixing this bug by shift right to one position.\n\nSigned-off-by: Eugenia Emantayev \u003ceugenia@mellanox.co.il\u003e\nReviewed-by: Yevgeny Petrilin \u003cyevgenyp@mellanox.co.il\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "33b5d30cd8b200024d55ac5830c9efde1102ef4c", "tree": "4788d50ad6994a40bda8e146db4ca9674a482b92", "parents": [ "3013dc0cceb9baaf25d5624034eeaa259bf99004", "c27111e5b8d6b1e9296f7075cdc35872f672b524" ], "author": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Wed Feb 15 13:41:52 2012 -0500" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Wed Feb 15 13:41:52 2012 -0500" }, "message": "Merge branch \u0027master\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless into for-davem\n" }, { "commit": "61cddc57dc14a5dffa0921d9a24fd68edbb374ac", "tree": "f909dda9bab29b39cecbd67bc9af008a0cf1ec03", "parents": [ "d65b4e98d7ea3038b767b70fe8be959b2913f16d" ], "author": { "name": "Lars-Peter Clausen", "email": "lars@metafoo.de", "time": "Wed Feb 15 10:23:25 2012 +0100" }, "committer": { "name": "Mark Brown", "email": "broonie@opensource.wolfsonmicro.com", "time": "Wed Feb 15 08:31:32 2012 -0800" }, "message": "regmap: Fix cache defaults initialization from raw cache defaults\n\nCurrently registers with a value of 0 are ignored when initializing the register\ndefaults from raw defaults. This worked in the past, because registers without a\nexplicit default were assumed to have a default value of 0. This was changed in\ncommit b03622a8 (\"regmap: Ensure rbtree syncs registers set to zero properly\").\nAs a result registers, which have a raw default value of 0 are now assumed to\nhave no default. This again can result in unnecessary writes when syncing the\ncache. It will also result in unnecessary reads for e.g. the first update\noperation. In the case where readback is not possible this will even let the\nupdate operation fail, if the register has not been written to before.\n\nSo this patch removes the check. Instead it adds a check to ignore raw defaults\nfor registers which are volatile, since those registers are not cached.\n\nSigned-off-by: Lars-Peter Clausen \u003clars@metafoo.de\u003e\nSigned-off-by: Mark Brown \u003cbroonie@opensource.wolfsonmicro.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "c38e23456278e967f094b08247ffc3711b1029b2", "tree": "ecd87e1fcd9c4e397d531081e45718cae132f7c1", "parents": [ "7ada1dd62804ca9ce1cb8666c6e563cd92fa50c1" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 15 08:05:18 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 15 08:05:18 2012 -0800" }, "message": "i387: fix sense of sanity check\n\nThe check for save_init_fpu() (introduced in commit 5b1cbac37798: \"i387:\nmake irq_fpu_usable() tests more robust\") was the wrong way around, but\nI hadn\u0027t noticed, because my \"tests\" were bogus: the FPU exceptions are\ndisabled by default, so even doing a divide by zero never actually\ntriggers this code at all unless you do extra work to enable them.\n\nSo if anybody did enable them, they\u0027d get one spurious warning.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6e2e340b59d2d4e7b6b7f2c2d02b0d5ca4df6458", "tree": "2fdbaf8bf120343c04a0db51eacca3d4b0ee0b19", "parents": [ "08a183f02b5fef1cd78d27ffc8281fa96d79f814" ], "author": { "name": "Tony Lindgren", "email": "tony@atomide.com", "time": "Tue Feb 14 21:58:56 2012 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Feb 15 11:04:36 2012 +0000" }, "message": "ARM: 7324/1: modpost: Fix section warnings for ARM for many compilers\n\nIt turns out that many compilers don\u0027t show section warnings on ARM\ncurrently because handling for ARM_CALL relocs are missing from\nmodpost.c.\n\nBased on commit c2e26114 ([ARM] 3205/1: Handle new EABI relocations when\nloading kernel modules) it seems that R_ARM_PC24, R_ARM_CALL and\nR_ARM_JUMP24 can be handled the same way.\n\nNote that at least Debian libc6-dev is missing defines for both\nR_ARM_CALL and R_ARM_JUMP24 in /usr/include/elf.h. So for now\nwe need to define them in modpost.c if not defined.\n\nAcked-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nCc: Ben Hutchings \u003cben@decadent.org.uk\u003e\nCc: Anders Kaseorg \u003candersk@ksplice.com\u003e\nCc: Greg KH \u003cgregkh@linuxfoundation.org\u003e\nSigned-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "08a183f02b5fef1cd78d27ffc8281fa96d79f814", "tree": "37621a2e6d7fe0cfe85465cef425467d33767ff1", "parents": [ "7ada1dd62804ca9ce1cb8666c6e563cd92fa50c1" ], "author": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Tue Feb 14 16:33:27 2012 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Feb 15 11:04:36 2012 +0000" }, "message": "ARM: 7323/1: Do not allow ARM_LPAE on pre-ARMv7 architectures\n\nThis patch expands the Kconfig dependencies for ARM_LPAE to not allow\nenabling when architectures other than ARMv7 are built into the kernel.\n\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nReported-by: Russell King \u003clinux@arm.linux.org.uk\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "7ada1dd62804ca9ce1cb8666c6e563cd92fa50c1", "tree": "210489b69f5db64fb0450a310ce9f8762121a66f", "parents": [ "ebf4bcbd5f249afc8f48a7aa32255f8ebbff5876", "9fc886a1880766fb91cb27c2e96bd814cef640d6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 14 15:26:42 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 14 15:26:42 2012 -0800" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator\n\nOne small bug fix from Axel plus a fix for a build failure in unrealistic\nbut commonly built configs which for some reason manage to survive for\nan awfully long time in -next without any reports.\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:\n regulator: Fix getting voltage in max8649_enable_time()\n regulator: Fix mc13xxx regulator modular build (again)\n" }, { "commit": "ebf4bcbd5f249afc8f48a7aa32255f8ebbff5876", "tree": "47fdb196fc109411d4524bf2c25d88a6d5470477", "parents": [ "8b36ac50da8e12d0469a7268ddbd6cab29307deb", "778a785f02ad846446e91dab49331bd7d853c514" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 14 15:21:25 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 14 15:21:25 2012 -0800" }, "message": "Merge branch \u0027merge\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc\n\nQuoth BenH:\n \"Here are a few powerpc fixes for 3.3, all pretty trivial. I also\n added the patch to define GET_IP/SET_IP so we can use some more\n asm-generic goodness.\"\n\n* \u0027merge\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:\n powerpc/pseries/eeh: Fix crash when error happens during device probe\n powerpc/pseries: Fix partition migration hang in stop_topology_update\n powerpc/powernv: Disable interrupts while taking phb-\u003elock\n powerpc: Fix WARN_ON in decrementer_check_overflow\n powerpc/wsp: Fix IRQ affinity setting\n powerpc: Implement GET_IP/SET_IP\n powerpc/wsp: Permanently enable PCI class code workaround\n" }, { "commit": "8b36ac50da8e12d0469a7268ddbd6cab29307deb", "tree": "4a48e1ea476dc4753f139eac01be0c4c576810fc", "parents": [ "694ce18ec3485f5b0a4f3820073f5e2b0b4e0870", "f9c2a0dc42a6938ff2a80e55ca2bbd1d5581c72e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 14 15:20:50 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 14 15:20:50 2012 -0800" }, "message": "Merge tag \u0027mmc-fixes-for-3.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc\n\nMMC fixes for 3.3-rc4:\n * The most visible fix here is against a regression introduced in 3.3-rc1\n that ran cards in Ultra High Speed mode even when they failed to initialize\n in that mode, leading to lower-speed cards failing to mount.\n * A lockdep warning introduced in 3.3-rc1 is fixed.\n * Various other small driver fixes, most notably for a NULL dereference\n when using highmem with dw_mmc.\n\n* tag \u0027mmc-fixes-for-3.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc:\n mmc: dw_mmc: Fix PIO mode with support of highmem\n mmc: atmel-mci: save and restore sdioirq when soft reset is performed\n mmc: block: Init ro_lock sysfs attr to fix lockdep warnings\n mmc: sh_mmcif: fix late delayed work initialisation\n mmc: tmio_mmc: fix card eject during IO with DMA\n mmc: core: Fix comparison issue in mmc_compare_ext_csds\n mmc: core: Fix PowerOff Notify suspend/resume\n mmc: sdhci-pci: set Medfield SDIO as non-removable\n mmc: core: add the capability for broken voltage\n mmc: core: Fix low speed mmc card detection failure\n mmc: esdhc: set the timeout to the max value\n mmc: esdhc: add PIO mode support\n mmc: core: Ensure clocks are always enabled before host interaction\n mmc: of_mmc_spi: fix little endian support\n mmc: core: UHS sdio card that fails should not exceed 50MHz\n mmc: esdhc: fix errors when booting kernel on Freescale eSDHC version 2.3\n" }, { "commit": "694ce18ec3485f5b0a4f3820073f5e2b0b4e0870", "tree": "e979926df34b41e17024107ee4731739608d8b51", "parents": [ "13d261932bbfff7f45f288c5c8cce43177cccd3b", "a43a5ccdfa5bd5b2f00aa9b2321df268c2e5d6e2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 14 15:20:11 2012 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 14 15:20:11 2012 -0800" }, "message": "Merge tag \u0027stable/for-linus-fixes-3.3-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen\n\nTwo fixes for VCPU offlining; One to fix the string format exposed\nby the xen-pci[front|back] to conform to the one used in majority of\nPCI drivers; Two fixes to make the code more resilient to invalid\nconfigurations.\n\nSigned-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n\n* tag \u0027stable/for-linus-fixes-3.3-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:\n xenbus_dev: add missing error check to watch handling\n xen/pci[front|back]: Use %d instead of %1x for displaying PCI devfn.\n xen pvhvm: do not remap pirqs onto evtchns if !xen_have_vector_callback\n xen/smp: Fix CPU online/offline bug triggering a BUG: scheduling while atomic.\n xen/bootup: During bootup suppress XENBUS: Unable to read cpu state\n" }, { "commit": "b9e44fe5ecda4158c22bc1ea4bffa378a4f83f65", "tree": "e6b5d7c69cd56c2f0f61475abfb89bc310ad4547", "parents": [ "887aa58e482dbf6d35d24cfa7b3315c8f5543339" ], "author": { "name": "li.rui27@zte.com.cn", "email": "li.rui27@zte.com.cn", "time": "Tue Feb 14 10:35:01 2012 +0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Feb 14 15:08:40 2012 -0800" }, "message": "USB: option: cleanup zte 3g-dongle\u0027s pid in option.c\n\n 1. Remove all old mass-storage ids\u0027s pid:\n 0x0026,0x0053,0x0098,0x0099,0x0149,0x0150,0x0160;\n 2. As the pid from 0x1401 to 0x1510 which have not surely assigned to\n use for serial-port or mass-storage port,so i think it should be\n removed now, and will re-add after it have assigned in future;\n 3. sort the pid to WCDMA and CDMA.\n\nSigned-off-by: Rui li \u003cli.rui27@zte.com.cn\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "887aa58e482dbf6d35d24cfa7b3315c8f5543339", "tree": "b72ffb8b5cb63f433afb59e7af4da0b92b9a2c86", "parents": [ "b7bc5925a9aafae970c3db1f118c97a7d503114e", "68d07f64b8a11a852d48d1b05b724c3e20c0d94b" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Feb 14 14:48:15 2012 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Feb 14 14:48:15 2012 -0800" }, "message": "Merge tag \u0027for-usb-linus-2012-02-14\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/sarah/xhci into usb-linus\n\nxHCI bug fix for broken BIOS on some Intel systems.\n\nHappy Valentine\u0027s Day! Your present is a working xHCI driver under some\nIntel systems with buggy BIOSes that don\u0027t provide a legacy PCI IRQ line\nfor the USB 3.0 host.\n\nPlease queue for 3.3 and stable to share the love!\n" }, { "commit": "3013dc0cceb9baaf25d5624034eeaa259bf99004", "tree": "2e02935bf2ddbcfc3f398484228b146000ef8eb9", "parents": [ "58e05f357a039a94aa36475f8c110256f693a239" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Tue Feb 14 10:27:09 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 14 16:28:23 2012 -0500" }, "message": "3c59x: shorten timer period for slave devices\n\nJean Delvare reported bonding on top of 3c59x adapters was not detecting\nnetwork cable removal fast enough.\n\n3c59x indeed uses a 60 seconds timer to check link status if carrier is\non, and 5 seconds if carrier is off.\n\nThis patch reduces timer period to 5 seconds if device is a bonding\nslave.\n\nReported-by: Jean Delvare \u003cjdelvare@suse.de\u003e\nAcked-by: Jean Delvare \u003cjdelvare@suse.de\u003e\nAcked-by: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "58e05f357a039a94aa36475f8c110256f693a239", "tree": "c9f03ac6514a6e4de646880be1f2e6d7764b8fea", "parents": [ "f65bd5ec47a4461bc575d5d34902fd18b6ec5542" ], "author": { "name": "Eric Dumazet", "email": "eric.dumazet@gmail.com", "time": "Tue Feb 14 10:11:59 2012 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 14 15:24:26 2012 -0500" }, "message": "netpoll: netpoll_poll_dev() should access dev-\u003eflags\n\ncommit 5a698af53f (bond: service netpoll arp queue on master device)\ntested IFF_SLAVE flag against dev-\u003epriv_flags instead of dev-\u003eflags\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nCc: WANG Cong \u003camwang@redhat.com\u003e\nAcked-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" } ], "next": "f65bd5ec47a4461bc575d5d34902fd18b6ec5542" }