)]}' { "log": [ { "commit": "62421645bb702c077ee5a462815525106cb53bcf", "tree": "83d98b98348bcaa8c7c0ddf2cc720104fdb6e52d", "parents": [ "9c06a46f1524a05b2dccf5c74d7555b2cfcbe5db" ], "author": { "name": "Phillip Lougher", "email": "phillip@squashfs.org.uk", "time": "Thu Nov 27 18:48:44 2014 +0000" }, "committer": { "name": "Phillip Lougher", "email": "phillip@squashfs.org.uk", "time": "Thu Nov 27 18:48:44 2014 +0000" }, "message": "Squashfs: Add LZ4 compression configuration option\n\nAdd the glue code, and also update the documentation.\n\nSigned-off-by: Phillip Lougher \u003cphillip@squashfs.org.uk\u003e\n" }, { "commit": "9c06a46f1524a05b2dccf5c74d7555b2cfcbe5db", "tree": "757488bc834c033f8bd491ff84adcc74fdc7e0e5", "parents": [ "5d01410fe4d92081f349b013a2e7a95429e4f2c9" ], "author": { "name": "Phillip Lougher", "email": "phillip@squashfs.org.uk", "time": "Thu Nov 27 06:49:14 2014 +0000" }, "committer": { "name": "Phillip Lougher", "email": "phillip@squashfs.org.uk", "time": "Thu Nov 27 07:44:11 2014 +0000" }, "message": "Squashfs: add LZ4 compression support\n\nAdd support for reading file systems compressed with the\nLZ4 compression algorithm.\n\nThis patch adds the LZ4 decompressor wrapper code.\n\nSigned-off-by: Phillip Lougher \u003cphillip@squashfs.org.uk\u003e\n" }, { "commit": "5d01410fe4d92081f349b013a2e7a95429e4f2c9", "tree": "333ee5d2b5031dc86d2f3c9d5a4bb4797bde9263", "parents": [ "82975bc6a6df743b9a01810fb32cb65d0ec5d60b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 15:25:20 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 15:25:20 2014 -0800" }, "message": "Linux 3.18-rc6\n" }, { "commit": "82975bc6a6df743b9a01810fb32cb65d0ec5d60b", "tree": "b5a5b092dcd8420d9f72bf66be2e6544a5622040", "parents": [ "90e362f4a75d0911ca75e5cd95591a6cf1f169dc" ], "author": { "name": "Andy Lutomirski", "email": "luto@amacapital.net", "time": "Fri Nov 21 13:26:07 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 14:25:28 2014 -0800" }, "message": "uprobes, x86: Fix _TIF_UPROBE vs _TIF_NOTIFY_RESUME\n\nx86 call do_notify_resume on paranoid returns if TIF_UPROBE is set but\nnot on non-paranoid returns. I suspect that this is a mistake and that\nthe code only works because int3 is paranoid.\n\nSetting _TIF_NOTIFY_RESUME in the uprobe code was probably a workaround\nfor the x86 bug. With that bug fixed, we can remove _TIF_NOTIFY_RESUME\nfrom the uprobes code.\n\nReported-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nAcked-by: Srikar Dronamraju \u003csrikar@linux.vnet.ibm.com\u003e\nAcked-by: Borislav Petkov \u003cbp@suse.de\u003e\nSigned-off-by: Andy Lutomirski \u003cluto@amacapital.net\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "90e362f4a75d0911ca75e5cd95591a6cf1f169dc", "tree": "c4a68a963a1da57469edec68db844f6e2fdda9b6", "parents": [ "00c89b2f1111b61e924f49fc79b7d9851fce249d" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Sun Nov 23 23:04:52 2014 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 14:14:40 2014 -0800" }, "message": "sched: Provide update_curr callbacks for stop/idle scheduling classes\n\nChris bisected a NULL pointer deference in task_sched_runtime() to\ncommit 6e998916dfe3 \u0027sched/cputime: Fix clock_nanosleep()/clock_gettime()\ninconsistency\u0027.\n\nChris observed crashes in atop or other /proc walking programs when he\nstarted fork bombs on his machine. He assumed that this is a new exit\nrace, but that does not make any sense when looking at that commit.\n\nWhat\u0027s interesting is that, the commit provides update_curr callbacks\nfor all scheduling classes except stop_task and idle_task.\n\nWhile nothing can ever hit that via the clock_nanosleep() and\nclock_gettime() interfaces, which have been the target of the commit in\nquestion, the author obviously forgot that there are other code paths\nwhich invoke task_sched_runtime()\n\ndo_task_stat(()\n thread_group_cputime_adjusted()\n thread_group_cputime()\n task_cputime()\n task_sched_runtime()\n if (task_current(rq, p) \u0026\u0026 task_on_rq_queued(p)) {\n update_rq_clock(rq);\n up-\u003esched_class-\u003eupdate_curr(rq);\n }\n\nIf the stats are read for a stomp machine task, aka \u0027migration/N\u0027 and\nthat task is current on its cpu, this will happily call the NULL pointer\nof stop_task-\u003eupdate_curr. Ooops.\n\nChris observation that this happens faster when he runs the fork bomb\nmakes sense as the fork bomb will kick migration threads more often so\nthe probability to hit the issue will increase.\n\nAdd the missing update_curr callbacks to the scheduler classes stop_task\nand idle_task. While idle tasks cannot be monitored via /proc we have\nother means to hit the idle case.\n\nFixes: 6e998916dfe3 \u0027sched/cputime: Fix clock_nanosleep()/clock_gettime() inconsistency\u0027\nReported-by: Chris Mason \u003cclm@fb.com\u003e\nReported-and-tested-by: Borislav Petkov \u003cbp@alien8.de\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Stanislaw Gruszka \u003csgruszka@redhat.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "00c89b2f1111b61e924f49fc79b7d9851fce249d", "tree": "4208da228a945babc6f07ee0536bd3ff66bc4503", "parents": [ "27946315d28cb8d1ea02321c4c673b1428d9315b", "b645af2d5905c4e32399005b867987919cbfc3ae" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 13:56:55 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 13:56:55 2014 -0800" }, "message": "Merge branch \u0027x86-traps\u0027 (trap handling from Andy Lutomirski)\n\nMerge x86-64 iret fixes from Andy Lutomirski:\n \"This addresses the following issues:\n\n - an unrecoverable double-fault triggerable with modify_ldt.\n - invalid stack usage in espfix64 failed IRET recovery from IST\n context.\n - invalid stack usage in non-espfix64 failed IRET recovery from IST\n context.\n\n It also makes a good but IMO scary change: non-espfix64 failed IRET\n will now report the correct error. Hopefully nothing depended on the\n old incorrect behavior, but maybe Wine will get confused in some\n obscure corner case\"\n\n* emailed patches from Andy Lutomirski \u003cluto@amacapital.net\u003e:\n x86_64, traps: Rework bad_iret\n x86_64, traps: Stop using IST for #SS\n x86_64, traps: Fix the espfix64 #DF fixup and rewrite it in C\n" }, { "commit": "b645af2d5905c4e32399005b867987919cbfc3ae", "tree": "63484dfff842bb500e4aaf47f7f50df830fbfe03", "parents": [ "6f442be2fb22be02cafa606f1769fa1e6f894441" ], "author": { "name": "Andy Lutomirski", "email": "luto@amacapital.net", "time": "Sat Nov 22 18:00:33 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 13:56:19 2014 -0800" }, "message": "x86_64, traps: Rework bad_iret\n\nIt\u0027s possible for iretq to userspace to fail. This can happen because\nof a bad CS, SS, or RIP.\n\nHistorically, we\u0027ve handled it by fixing up an exception from iretq to\nland at bad_iret, which pretends that the failed iret frame was really\nthe hardware part of #GP(0) from userspace. To make this work, there\u0027s\nan extra fixup to fudge the gs base into a usable state.\n\nThis is suboptimal because it loses the original exception. It\u0027s also\nbuggy because there\u0027s no guarantee that we were on the kernel stack to\nbegin with. For example, if the failing iret happened on return from an\nNMI, then we\u0027ll end up executing general_protection on the NMI stack.\nThis is bad for several reasons, the most immediate of which is that\ngeneral_protection, as a non-paranoid idtentry, will try to deliver\nsignals and/or schedule from the wrong stack.\n\nThis patch throws out bad_iret entirely. As a replacement, it augments\nthe existing swapgs fudge into a full-blown iret fixup, mostly written\nin C. It\u0027s should be clearer and more correct.\n\nSigned-off-by: Andy Lutomirski \u003cluto@amacapital.net\u003e\nReviewed-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6f442be2fb22be02cafa606f1769fa1e6f894441", "tree": "1173ca6feb0cdda19b5383e713e2f19152d88940", "parents": [ "af726f21ed8af2cdaa4e93098dc211521218ae65" ], "author": { "name": "Andy Lutomirski", "email": "luto@amacapital.net", "time": "Sat Nov 22 18:00:32 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 13:56:19 2014 -0800" }, "message": "x86_64, traps: Stop using IST for #SS\n\nOn a 32-bit kernel, this has no effect, since there are no IST stacks.\n\nOn a 64-bit kernel, #SS can only happen in user code, on a failed iret\nto user space, a canonical violation on access via RSP or RBP, or a\ngenuine stack segment violation in 32-bit kernel code. The first two\ncases don\u0027t need IST, and the latter two cases are unlikely fatal bugs,\nand promoting them to double faults would be fine.\n\nThis fixes a bug in which the espfix64 code mishandles a stack segment\nviolation.\n\nThis saves 4k of memory per CPU and a tiny bit of code.\n\nSigned-off-by: Andy Lutomirski \u003cluto@amacapital.net\u003e\nReviewed-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "af726f21ed8af2cdaa4e93098dc211521218ae65", "tree": "d3546c0fe44e49d283ed3d0ced3b92873f28df34", "parents": [ "fc14f9c1272f62c3e8d01300f52467c0d9af50f9" ], "author": { "name": "Andy Lutomirski", "email": "luto@amacapital.net", "time": "Sat Nov 22 18:00:31 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 13:56:18 2014 -0800" }, "message": "x86_64, traps: Fix the espfix64 #DF fixup and rewrite it in C\n\nThere\u0027s nothing special enough about the espfix64 double fault fixup to\njustify writing it in assembly. Move it to C.\n\nThis also fixes a bug: if the double fault came from an IST stack, the\nold asm code would return to a partially uninitialized stack frame.\n\nFixes: 3891a04aafd668686239349ea58f3314ea2af86b\nSigned-off-by: Andy Lutomirski \u003cluto@amacapital.net\u003e\nReviewed-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "27946315d28cb8d1ea02321c4c673b1428d9315b", "tree": "74b23a4719ac86d0a2d292abe4d65c01e099bc95", "parents": [ "9f2e0f6370b5f37cea3deee47299c8dfb0228eeb", "49e41938f8a53d01d1c5d133a038acb9650008f9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 11:46:01 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 11:46:01 2014 -0800" }, "message": "Merge tag \u0027armsoc-for-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc\n\nPull ARM SoC fixes from Olof Johansson:\n \"A collection of fixes this week:\n\n - A set of clock fixes for shmobile platforms\n - A fix for tegra that moves serial port labels to be per board.\n We\u0027re choosing to merge this for 3.18 because the labels will start\n being parsed in 3.19, and without this change serial port numbers\n that used to be stable since the dawn of time will change numbers.\n - A few other DT tweaks for Tegra.\n - A fix for multi_v7_defconfig that makes it stop spewing cpufreq\n errors on Arndale (Exynos)\"\n\n* tag \u0027armsoc-for-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:\n ARM: multi_v7_defconfig: fix failure setting CPU voltage by enabling dependent I2C controller\n ARM: tegra: roth: Fix SD card VDD_IO regulator\n ARM: tegra: Remove eMMC vmmc property for roth/tn7\n ARM: dts: tegra: move serial aliases to per-board\n ARM: tegra: Add serial port labels to Tegra124 DT\n ARM: shmobile: kzm9g legacy: Set i2c clks_per_count to 2\n ARM: shmobile: r8a7740 dtsi: Correct IIC0 parent clock\n ARM: shmobile: r8a7790: Fix SD3CKCR address to device tree\n ARM: shmobile: r8a7740 legacy: Correct IIC0 parent clock\n ARM: shmobile: r8a7740 legacy: Add missing INTCA clock for irqpin module\n ARM: shmobile: r8a7790: Fix SD3CKCR address\n ARM: dts: sun6i: Re-parent ahb1_mux to pll6 as required by dma controller\n" }, { "commit": "9f2e0f6370b5f37cea3deee47299c8dfb0228eeb", "tree": "1f7d684f6e84876806fa1f3cc5fab7029d7cf0d3", "parents": [ "d038a63ace6cf2ce3aeafa741b73d542ffb65163", "4aab3b5b3ccf94fc907e66233e6ca4d8675759a6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 11:33:49 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 11:33:49 2014 -0800" }, "message": "Merge branch \u0027for-3.18-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu\n\nPull percpu fix from Tejun Heo:\n \"This contains one patch to fix a race condition which can lead to\n percpu_ref using a percpu pointer which is corrupted with a set DEAD\n bit. The bug was introduced while separating out the ATOMIC mode flag\n from the DEAD flag. The fix is pretty straight forward.\n\n I just committed the patch to the percpu tree but am sending out the\n pull request early as I\u0027ll be on vacation for a week. The patch\n should be fairly safe and while the latency will be higher I\u0027ll be\n checking emails\"\n\n* \u0027for-3.18-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu:\n percpu-ref: fix DEAD flag contamination of percpu pointer\n" }, { "commit": "d038a63ace6cf2ce3aeafa741b73d542ffb65163", "tree": "9d7b2bcc203fb3074a1f768fa30b071042690013", "parents": [ "cb95413971d605b0d152d3ceecc47ba8991d66fb", "f82c458a2c3ffb94b431fc6ad791a79df1b3713e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 11:16:36 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 23 11:16:36 2014 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs\n\nPull btrfs deadlock fix from Chris Mason:\n \"This has a fix for a long standing deadlock that we\u0027ve been trying to\n nail down for a while. It ended up being a bad interaction with the\n fair reader/writer locks and the order btrfs reacquires locks in the\n btree\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:\n btrfs: fix lockups from btrfs_clear_path_blocking\n" }, { "commit": "4aab3b5b3ccf94fc907e66233e6ca4d8675759a6", "tree": "c14d2a47ec7cd23e11525ae608c88433c392ed01", "parents": [ "cac7f2429872d3733dc3f9915857b1691da2eb2f" ], "author": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Sat Nov 22 09:22:42 2014 -0500" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Sun Nov 23 12:36:06 2014 -0500" }, "message": "percpu-ref: fix DEAD flag contamination of percpu pointer\n\nWhile decoupling ATOMIC and DEAD flags, f47ad4578461 (\"percpu_ref:\ndecouple switching to percpu mode and reinit\") updated\n__ref_is_percpu() so that it only tests ATOMIC flag to determine\nwhether the ref is in percpu mode or not; however, while DEAD implies\nATOMIC, the two flags are set separately during percpu_ref_kill() and\nif __ref_is_percpu() races percpu_ref_kill(), it may see DEAD w/o\nATOMIC. Because __ref_is_percpu() returns @ref-\u003epercpu_count_ptr\nvalue verbatim as the percpu pointer after testing ATOMIC, the pointer\nmay now be contaminated with the DEAD flag.\n\nThis can be fixed by clearing the flag bits before returning the\npointer which was the fix proposed by Shaohua; however, as DEAD\nimplies ATOMIC, we can just test for both flags at once and avoid the\nexplicit masking.\n\nUpdate __ref_is_percpu() so that it tests that both ATOMIC and DEAD\nare clear before returning @ref-\u003epercpu_count_ptr as the percpu\npointer.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nReported-and-Reviewed-by: Shaohua Li \u003cshli@kernel.org\u003e\nLink: http://lkml.kernel.org/r/995deb699f5b873c45d667df4add3b06f73c2c25.1416638887.git.shli@kernel.org\nFixes: f47ad4578461 (\"percpu_ref: decouple switching to percpu mode and reinit\")\n" }, { "commit": "cb95413971d605b0d152d3ceecc47ba8991d66fb", "tree": "82d8fa9cd7f34049bbef455562488ea4c96c75a9", "parents": [ "ecde00642c79c2dfc9006e7751ee60bd0e21e1e9", "6bab4a8a1888729f17f4923cc5867e4674f66333" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 22 14:33:11 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 22 14:33:11 2014 -0800" }, "message": "Merge branch \u0027timers-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull timer fix from Thomas Gleixner:\n \"A single bugfix for an init order problem in the sun4i subarch\n clockevents code\"\n\n* \u0027timers-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n clockevent: sun4i: Fix race condition in the probe code\n" }, { "commit": "ecde00642c79c2dfc9006e7751ee60bd0e21e1e9", "tree": "de4cca84e38a44c46a0d8f82b47aa96dba44c8e9", "parents": [ "8a84e01e147f44111988f9d8ccd2eaa30215a0f2", "3035b675ad0377b2468b442068062e7f28771dda" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 22 14:15:27 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 22 14:15:27 2014 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull vfs fixes from Al Viro:\n \"Assorted fixes, most in overlayfs land\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:\n ovl: ovl_dir_fsync() cleanup\n ovl: update MAINTAINERS\n ovl: pass dentry into ovl_dir_read_merged()\n ovl: use lockless_dereference() for upperdentry\n ovl: allow filenames with comma\n ovl: fix race in private xattr checks\n ovl: fix remove/copy-up race\n ovl: rename filesystem type to \"overlay\"\n isofs: avoid unused function warning\n vfs: fix reference leak in d_prune_aliases()\n" }, { "commit": "8a84e01e147f44111988f9d8ccd2eaa30215a0f2", "tree": "77617b3ab40f5c97575400db348b120c36871ed8", "parents": [ "928352e9eebabb814d0c38af1772af55677faf62", "0c228e833c88e3aa029250f5db77d5968c5ce5b5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 17:20:36 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 17:20:36 2014 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net\n\nPull networking fixes from David Miller:\n\n 1) Fix BUG when decrypting empty packets in mac80211, from Ronald Wahl.\n\n 2) nf_nat_range is not fully initialized and this is copied back to\n userspace, from Daniel Borkmann.\n\n 3) Fix read past end of b uffer in netfilter ipset, also from Dan\n Carpenter.\n\n 4) Signed integer overflow in ipv4 address mask creation helper\n inet_make_mask(), from Vincent BENAYOUN.\n\n 5) VXLAN, be2net, mlx4_en, and qlcnic need -\u003endo_gso_check() methods to\n properly describe the device\u0027s capabilities, from Joe Stringer.\n\n 6) Fix memory leaks and checksum miscalculations in openvswitch, from\n Pravin B SHelar and Jesse Gross.\n\n 7) FIB rules passes back ambiguous error code for unreachable routes,\n making behavior confusing for userspace. Fix from Panu Matilainen.\n\n 8) ieee802154fake_probe() doesn\u0027t release resources properly on error,\n from Alexey Khoroshilov.\n\n 9) Fix skb_over_panic in add_grhead(), from Daniel Borkmann.\n\n10) Fix access of stale slave pointers in bonding code, from Nikolay\n Aleksandrov.\n\n11) Fix stack info leak in PPP pptp code, from Mathias Krause.\n\n12) Cure locking bug in IPX stack, from Jiri Bohac.\n\n13) Revert SKB fclone memory freeing optimization that is racey and can\n allow accesses to freed up memory, from Eric Dumazet.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (71 commits)\n tcp: Restore RFC5961-compliant behavior for SYN packets\n net: Revert \"net: avoid one atomic operation in skb_clone()\"\n virtio-net: validate features during probe\n cxgb4 : Fix DCB priority groups being returned in wrong order\n ipx: fix locking regression in ipx_sendmsg and ipx_recvmsg\n openvswitch: Don\u0027t validate IPv6 label masks.\n pptp: fix stack info leak in pptp_getname()\n brcmfmac: don\u0027t include linux/unaligned/access_ok.h\n cxgb4i : Don\u0027t block unload/cxgb4 unload when remote closes TCP connection\n ipv6: delete protocol and unregister rtnetlink when cleanup\n net/mlx4_en: Add VXLAN ndo calls to the PF net device ops too\n bonding: fix curr_active_slave/carrier with loadbalance arp monitoring\n mac80211: minstrel_ht: fix a crash in rate sorting\n vxlan: Inline vxlan_gso_check().\n can: m_can: update to support CAN FD features\n can: m_can: fix incorrect error messages\n can: m_can: add missing delay after setting CCCR_INIT bit\n can: m_can: fix not set can_dlc for remote frame\n can: m_can: fix possible sleep in napi poll\n can: m_can: add missing message RAM initialization\n ...\n" }, { "commit": "928352e9eebabb814d0c38af1772af55677faf62", "tree": "86e2d3f998a25fb72a1197e81ba9e31ad2a3b9e7", "parents": [ "9a7e4f5633f0c733820091cc9c643cc0c257c349", "a0fc608178a9b38a5f782331909fcc208b742a7b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 17:15:28 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 17:15:28 2014 -0800" }, "message": "Merge branch \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull drm fixes from Dave Airlie:\n \"Just two radeon and two intel fixes: endian and regression fixes\"\n\n* \u0027drm-fixes\u0027 of git://people.freedesktop.org/~airlied/linux:\n drm/radeon: fix endian swapping in vbios fetch for tdp table\n drm/radeon: disable native backlight control on pre-r6xx asics (v2)\n drm/i915: Kick fbdev before vgacon\n drm/i915: drop WaSetupGtModeTdRowDispatch:snb\n" }, { "commit": "9a7e4f5633f0c733820091cc9c643cc0c257c349", "tree": "1da51394d6f3794fe5b4e645002222709e5f6c2b", "parents": [ "f100a74674786304a3142be49286dd96a74581a0", "6676f3081f7e3dae64e05b87d47a041b782f898a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 17:11:56 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 17:11:56 2014 -0800" }, "message": "Merge tag \u0027sound-3.18-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound\n\nPull sound fixes from Takashi Iwai:\n \"This batch ended up as a relatively high volume due to pending ASoC\n fixes. But most of fixes there are trivial and/or device- specific\n fixes and quirks, so safe to apply. The only (ASoC) core fixes are\n the DPCM race fix and the machine-driver matching fix for\n componentization\"\n\n* tag \u0027sound-3.18-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:\n ALSA: hda - fix the mic mute led problem for Latitude E5550\n ALSA: hda - move DELL_WMI_MIC_MUTE_LED to the tail in the quirk chain\n ASoC: wm_adsp: Avoid attempt to free buffers that might still be in use\n ALSA: usb-audio: Set the Control Selector to SU_SELECTOR_CONTROL for UAC2\n ALSA: usb-audio: Add ctrl message delay quirk for Marantz/Denon devices\n ASoC: sgtl5000: Fix SMALL_POP bit definition\n ASoC: cs42l51: re-hook of_match_table pointer\n ASoC: rt5670: change dapm routes of PLL connection\n ASoC: rt5670: correct the incorrect default values\n ASoC: samsung: Add MODULE_DEVICE_TABLE for Snow\n ASoC: max98090: Correct pclk divisor settings\n ASoC: dpcm: Fix race between FE/BE updates and trigger\n ASoC: Fix snd_soc_find_dai() matching component by name\n ASoC: rsnd: remove unsupported PAUSE flag\n ASoC: fsi: remove unsupported PAUSE flag\n ASoC: rt5645: Mark RT5645_TDM_CTRL_3 as readable\n ASoC: rockchip-i2s: fix infinite loop in rockchip_snd_rxctrl\n ASoC: es8328-i2c: Fix i2c_device_id name field in es8328_id\n ASoC: fsl_asrc: Add reg_defaults for regmap to fix kernel dump\n" }, { "commit": "f100a74674786304a3142be49286dd96a74581a0", "tree": "60237defc914d05529d3fdf09b388afce0bf0a4f", "parents": [ "2e29a6d086d25ece47a84111275ebb14baa054ef", "78579b7c7eb45f0e7ec5e9437087ed21749f9a9c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:56:25 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:56:25 2014 -0800" }, "message": "Merge tag \u0027pm+acpi-3.18-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull ACPI power management fix from Rafael Wysocki:\n \"This is just a one-liner fixing a regression introduced in 3.13 that\n broke system suspend on some Chromebooks.\n\n On those machines there are ACPI device objects for some I2C devices\n that can wake up the system from sleep states, but that is done via a\n platform-specific mechanism and the ACPI objects don\u0027t contain any\n wakeup-related information. When we started to use ACPI power\n management with those devices (which happened during the 3.13 cycle),\n their configuration confused the ACPI PM layer that returned error\n codes from suspend callbacks for them causing system suspend to fail.\n\n However, the ACPI PM layer can safely ignore the wakeup setting from a\n device driver if the ACPI object corresponding to the device in\n question doesn\u0027t contain wakeup information in which case the driver\n itself is responsible for setting up the device for system wakeup\"\n\n* tag \u0027pm+acpi-3.18-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n ACPI / PM: Ignore wakeup setting if the ACPI companion can\u0027t wake up\n" }, { "commit": "2e29a6d086d25ece47a84111275ebb14baa054ef", "tree": "a5afcc7e0b36b836f1bdbe2cbd28445394fd5250", "parents": [ "08685897b3586aad622cb48fe1fb07bc19bb78f5", "788ec2fc2ca295a2d929986e95231214ecd8d142" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:40:41 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:40:41 2014 -0800" }, "message": "Merge tag \u0027devicetree-fixes-for-3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux\n\nPull devicetree fixes from Rob Herring:\n \"DeviceTree fixes for 3.18:\n\n - two fixes for OF selftest code\n - fix for PowerPC address parsing to disable work-around except on\n old PowerMACs\n - fix a crash when earlycon is enabled, but no device is found\n - DT documentation fixes and missing vendor prefixes\n\n All but the doc updates are also for stable\"\n\n* tag \u0027devicetree-fixes-for-3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:\n of/selftest: Fix testing when /aliases is missing\n of/selftest: Fix off-by-one error in removal path\n documentation: pinctrl bindings: Fix trivial typo \u0027abitrary\u0027\n devicetree: bindings: Add vendor prefix for Micron Technology, Inc.\n of: Add vendor prefix for Chips\u0026Media, Inc.\n of/base: Fix PowerPC address parsing hack\n devicetree: vendor-prefixes.txt: fix whitespace\n of: Fix crash if an earlycon driver is not found\n of/irq: Drop obsolete \u0027interrupts\u0027 vs \u0027interrupts-extended\u0027 text\n of: Spelling s/stucture/structure/\n devicetree: bindings: add sandisk to the vendor prefixes\n" }, { "commit": "08685897b3586aad622cb48fe1fb07bc19bb78f5", "tree": "64eab4dfa420b498409aa24d3cb4182d08d81cb3", "parents": [ "a46171d0100eafc0c276962d80f470406d66dcdd", "7fc986d8a9727e5d40da3c2c1c343da6142e82a9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:36:42 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:36:42 2014 -0800" }, "message": "Merge tag \u0027pci-v3.18-fixes-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci\n\nPull PCI fixes from Bjorn Helgaas:\n \"These are fixes for an issue with 64-bit PCI bus addresses on 32-bit\n PAE kernels, an APM X-Gene problem (it depended on a generic change we\n removed before merging), a fix for my hotplug device configuration\n changes, and a devicetree documentation update.\n\n Resource management:\n - Support 64-bit bridge windows if we have 64-bit dma_addr_t (Yinghai Lu)\n\n PCI device hotplug:\n - Apply _HPX Link Control settings to all devices with a link (Yinghai Lu)\n\n Generic host bridge driver:\n - Add DT binding for \"linux,pci-domain\" property (Lucas Stach)\n\n APM X-Gene:\n - Assign resources to bus before adding new devices (Duc Dang)\"\n\n* tag \u0027pci-v3.18-fixes-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:\n PCI: Support 64-bit bridge windows if we have 64-bit dma_addr_t\n PCI: Apply _HPX Link Control settings to all devices with a link\n PCI: Add missing DT binding for \"linux,pci-domain\" property\n PCI: xgene: Assign resources to bus before adding new devices\n" }, { "commit": "a46171d0100eafc0c276962d80f470406d66dcdd", "tree": "ed2a94222068a7a4da1c19aeb8edc47ebe1f4d56", "parents": [ "4ec69c7ebc2cb8df1f33b26f4492481c452fef66", "b1a5ad006b34ded9dc7ec64988deba1b3ecad367" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:28:45 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:28:45 2014 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending\n\nPull SCSI target fixes from Nicholas Bellinger:\n \"Here are the target-pending fixes queued for v3.18-rc6.\n\n The highlights include:\n\n - target-core OOPs fix with tcm_qla2xxx + vxworks FC initiators +\n zero length SCSI commands having a transfer direction set. (Roland\n + Craig Watson)\n\n - vhost-scsi OOPs fix to explicitly prevent WWPN endpoint configfs\n group removal while qemu still has an active reference. (Paolo +\n nab)\n\n - ib_srpt fix for RDMA hardware with lower srp_sq_size limits.\n (Bart)\n\n - two ib_isert work-arounds for running on ocrdma hardware (Or + Sagi\n + Chris)\n\n - iscsi-target discovery portal typo + SPC-3 PR Preempt SA key\n matching fix (Steve)\"\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:\n IB/isert: Adjust CQ size to HW limits\n target: return CONFLICT only when SA key unmatched\n iser-target: Handle DEVICE_REMOVAL event on network portal listener correctly\n ib_isert: Add max_send_sge\u003d2 minimum for control PDU responses\n srp-target: Retry when QP creation fails with ENOMEM\n iscsi-target: return the correct port in SendTargets\n vhost-scsi: Take configfs group dependency during VHOST_SCSI_SET_ENDPOINT\n target: Don\u0027t call TFO-\u003ewrite_pending if data_length \u003d\u003d 0\n" }, { "commit": "4ec69c7ebc2cb8df1f33b26f4492481c452fef66", "tree": "be991a02923758c2afe09a13fe61650686bac8df", "parents": [ "e6a588d086a75dc20afb8ffbcbe23a50d4a1ca37", "1f0a5cbf61a54504236bbbe2c98b58e85f90e650" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:24:27 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:24:27 2014 -0800" }, "message": "Merge branch \u0027fixes\u0027 of git://git.infradead.org/users/vkoul/slave-dma\n\nPull dmaengine fixes from Vinod Koul:\n \"We have couple of fixes for dmaengine queued up:\n - dma mempcy fix for dma configuration of sun6i by Maxime\n - pl330 fixes: First the fixing allocation for data buffers by Liviu\n and then Jon\u0027s fixe for fifo width and usage\"\n\n* \u0027fixes\u0027 of git://git.infradead.org/users/vkoul/slave-dma:\n dmaengine: Fix allocation size for PL330 data buffer depth.\n dmaengine: pl330: Limit MFIFO usage for memcpy to avoid exhausting entries\n dmaengine: pl330: Align DMA memcpy operations to MFIFO width\n dmaengine: sun6i: Fix memcpy operation\n" }, { "commit": "e6a588d086a75dc20afb8ffbcbe23a50d4a1ca37", "tree": "ec8ca3767ca078b2f215b0239f021d0fe261e6a1", "parents": [ "4fc82c0a766cf1d0bc098fb42d00b5292dde65f7", "935c2dbec4d6d3163ee8e7409996904a734ad89a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:14:58 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:14:58 2014 -0800" }, "message": "Merge branch \u0027upstream\u0027 of git://git.linux-mips.org/pub/scm/ralf/upstream-linus\n\nPull MIPS fixes from Ralf Baechle:\n \"More 3.18 fixes for MIPS:\n\n - backtraces were not quite working on on 64-bit kernels\n - loongson needs a different cache coherency setting\n - Loongson 3 is a MIPS64 R2 version but due to erratum we treat is an\n older architecture revision.\n - fix build errors due to undefined references to __node_distances\n for certain configurations.\n - fix instruction decodig in the jump label code.\n - for certain configurations copy_{from,to}_user destroy the content\n of $3 so that register needs to be marked as clobbed by the calling\n code.\n - Hardware Table Walker fixes.\n - fill the delay slot of the last instruction of memcpy otherwise\n whatever ends up there randomly might have undesirable effects.\n - ensure get_user/__get_user always zero the variable to be read even\n in case of an error\"\n\n* \u0027upstream\u0027 of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:\n MIPS: jump_label.c: Handle the microMIPS J instruction encoding\n MIPS: jump_label.c: Correct the span of the J instruction\n MIPS: Zero variable read by get_user / __get_user in case of an error.\n MIPS: lib: memcpy: Restore NOP on delay slot before returning to caller\n MIPS: tlb-r4k: Add missing HTW stop/start sequences\n MIPS: asm: uaccess: Add v1 register to clobber list on EVA\n MIPS: oprofile: Fix backtrace on 64-bit kernel\n MIPS: Loongson: Set Loongson-3\u0027s ISA level to MIPS64R1\n MIPS: Loongson: Fix the write-combine CCA value setting\n MIPS: IP27: Fix __node_distances undefined error\n MIPS: Loongson3: Fix __node_distances undefined error\n" }, { "commit": "4fc82c0a766cf1d0bc098fb42d00b5292dde65f7", "tree": "8da4b82e6f7f3e40a88074a973392f987281b360", "parents": [ "c6c9161d064d30e78904f3affe5184487493e0fc", "d7ce4377494adfaf8afb15ecf4f07d399bbf13d9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:13:34 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 16:13:34 2014 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux\n\nPull powerpc fix from Michael Ellerman:\n \"One fix from Scott, he says:\n\n This patch fixes a crash (introduced in v3.18-rc1) in the FSL MSI driver\n when threaded IRQs are enabled\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux:\n powerpc/fsl_msi: mark the msi cascade handler IRQF_NO_THREAD\n" }, { "commit": "c6c9161d064d30e78904f3affe5184487493e0fc", "tree": "07387e6ee769c2cf24d13698cbe84f1af679a669", "parents": [ "8b2ed21e846c63d8f1bdee0d8df0645721a604a1", "b5e212a3051b65e426a513901d9c7001681c7215" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 15:46:17 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 15:46:17 2014 -0800" }, "message": "Merge branch \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Thomas Gleixner:\n \"Misc fixes:\n - gold linker build fix\n - noxsave command line parsing fix\n - bugfix for NX setup\n - microcode resume path bug fix\n - _TIF_NOHZ versus TIF_NOHZ bugfix as discussed in the mysterious\n lockup thread\"\n\n* \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86, syscall: Fix _TIF_NOHZ handling in syscall_trace_enter_phase1\n x86, kaslr: Handle Gold linker for finding bss/brk\n x86, mm: Set NX across entire PMD at boot\n x86, microcode: Update BSPs microcode on resume\n x86: Require exact match for \u0027noxsave\u0027 command line option\n" }, { "commit": "8b2ed21e846c63d8f1bdee0d8df0645721a604a1", "tree": "459906e25111254256e4873043da98a6e4055564", "parents": [ "13f5004c94785af107dd702d9fbbe160f1004064", "6e998916dfe327e785e7c2447959b2c1a3ea4930" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 15:44:54 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 15:44:54 2014 -0800" }, "message": "Merge branch \u0027sched-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull scheduler fixes from Ingo Molnar:\n \"Misc fixes: two NUMA fixes, two cputime fixes and an RCU/lockdep fix\"\n\n* \u0027sched-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n sched/cputime: Fix clock_nanosleep()/clock_gettime() inconsistency\n sched/cputime: Fix cpu_timer_sample_group() double accounting\n sched/numa: Avoid selecting oneself as swap target\n sched/numa: Fix out of bounds read in sched_init_numa()\n sched: Remove lockdep check in sched_move_task()\n" }, { "commit": "13f5004c94785af107dd702d9fbbe160f1004064", "tree": "4aded04e423a06fd232e653cb5efefa98f342773", "parents": [ "a64bb02f4a62a604d8dd62decb559b9c6adfb40c", "68055915c1c22489f9658bd2b7391bb11b2cf4e4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 15:44:07 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 15:44:07 2014 -0800" }, "message": "Merge branch \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf fixes from Ingo Molnar:\n \"Misc fixes: two Intel uncore driver fixes, a CPU-hotplug fix and a\n build dependencies fix\"\n\n* \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n perf/x86/intel/uncore: Fix boot crash on SBOX PMU on Haswell-EP\n perf/x86/intel/uncore: Fix IRP uncore register offsets on Haswell EP\n perf: Fix corruption of sibling list with hotplug\n perf/x86: Fix embarrasing typo\n" }, { "commit": "a64bb02f4a62a604d8dd62decb559b9c6adfb40c", "tree": "56d0f7ba555d5b62e4043464f320baf239c05cbe", "parents": [ "fc14f9c1272f62c3e8d01300f52467c0d9af50f9", "00b4d9a14125f1e51874def2b9de6092e007412d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 15:38:21 2014 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 21 15:38:21 2014 -0800" }, "message": "Merge branch \u0027core-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull core fix from Ingo Molnar:\n \"Fix GENMASK macro shift overflow\"\n\nNobody seems to currently use GENMASK() to fill every single last bit\n(which is what overflows) in-tree, and gcc would warn about it, so we\nhave that going for us. But apparently there are pending changes that\nwant this.\n\n* \u0027core-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n bitops: Fix shift overflow in GENMASK macros\n" }, { "commit": "0c228e833c88e3aa029250f5db77d5968c5ce5b5", "tree": "8197a1364b857bb53cde500b1db28eeb43dc6f9c", "parents": [ "e7820e39b7d19b9fe1928fc19de9361b44150ca6" ], "author": { "name": "Calvin Owens", "email": "calvinowens@fb.com", "time": "Thu Nov 20 15:09:53 2014 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 21 15:33:50 2014 -0500" }, "message": "tcp: Restore RFC5961-compliant behavior for SYN packets\n\nCommit c3ae62af8e755 (\"tcp: should drop incoming frames without ACK\nflag set\") was created to mitigate a security vulnerability in which a\nlocal attacker is able to inject data into locally-opened sockets by\nusing TCP protocol statistics in procfs to quickly find the correct\nsequence number.\n\nThis broke the RFC5961 requirement to send a challenge ACK in response\nto spurious RST packets, which was subsequently fixed by commit\n7b514a886ba50 (\"tcp: accept RST without ACK flag\").\n\nUnfortunately, the RFC5961 requirement that spurious SYN packets be\nhandled in a similar manner remains broken.\n\nRFC5961 section 4 states that:\n\n ... the handling of the SYN in the synchronized state SHOULD be\n performed as follows:\n\n 1) If the SYN bit is set, irrespective of the sequence number, TCP\n MUST send an ACK (also referred to as challenge ACK) to the remote\n peer:\n\n \u003cSEQ\u003dSND.NXT\u003e\u003cACK\u003dRCV.NXT\u003e\u003cCTL\u003dACK\u003e\n\n After sending the acknowledgment, TCP MUST drop the unacceptable\n segment and stop processing further.\n\n By sending an ACK, the remote peer is challenged to confirm the loss\n of the previous connection and the request to start a new connection.\n A legitimate peer, after restart, would not have a TCB in the\n synchronized state. Thus, when the ACK arrives, the peer should send\n a RST segment back with the sequence number derived from the ACK\n field that caused the RST.\n\n This RST will confirm that the remote peer has indeed closed the\n previous connection. Upon receipt of a valid RST, the local TCP\n endpoint MUST terminate its connection. The local TCP endpoint\n should then rely on SYN retransmission from the remote end to\n re-establish the connection.\n\nThis patch lets SYN packets through the discard added in c3ae62af8e755,\nso that spurious SYN packets are properly dealt with as per the RFC.\n\nThe challenge ACK is sent unconditionally and is rate-limited, so the\noriginal vulnerability is not reintroduced by this patch.\n\nSigned-off-by: Calvin Owens \u003ccalvinowens@fb.com\u003e\nAcked-by: Eric Dumazet \u003cedumazet@google.com\u003e\nAcked-by: Neal Cardwell \u003cncardwell@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e7820e39b7d19b9fe1928fc19de9361b44150ca6", "tree": "15599feec720efe469184ed07d9eefc91c32a086", "parents": [ "892d6eb1245b771987afb8667a65344e568d3439" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Fri Nov 21 11:47:16 2014 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 21 15:26:32 2014 -0500" }, "message": "net: Revert \"net: avoid one atomic operation in skb_clone()\"\n\nNot sure what I was thinking, but doing anything after\nreleasing a refcount is suicidal or/and embarrassing.\n\nBy the time we set skb-\u003efclone to SKB_FCLONE_FREE, another cpu\ncould have released last reference and freed whole skb.\n\nWe potentially corrupt memory or trap if CONFIG_DEBUG_PAGEALLOC is set.\n\nReported-by: Chris Mason \u003cclm@fb.com\u003e\nFixes: ce1a4ea3f1258 (\"net: avoid one atomic operation in skb_clone()\")\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nCc: Sabrina Dubroca \u003csd@queasysnail.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3035b675ad0377b2468b442068062e7f28771dda", "tree": "63cf26aa249d067ac12776dd2850ff017747f2b9", "parents": [ "7ca2f234404e738cc807ed87c43f9932513bc8c6", "7676895f4736421ebafc48de5078e25ea69e88ee" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri Nov 21 11:51:08 2014 -0500" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri Nov 21 11:51:08 2014 -0500" }, "message": "Merge branch \u0027overlayfs-current\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs into for-linus\n\n\"The biggest change is to rename the filesystem from \"overlayfs\" to \"overlay\".\nThis will allow legacy overlayfs to be easily carried by distros alongside the\nnew mainline one. Also fix a couple of copy-up races and allow escaping comma\ncharacter in filenames.\"\n\nThe last bit is about commas in pathname mount options...\n" }, { "commit": "892d6eb1245b771987afb8667a65344e568d3439", "tree": "5695d9a2cffaa2d0772d573e8a273f1a4555206d", "parents": [ "7e09dccd07518729fe3cf586beb83acffa2e64ca" ], "author": { "name": "Jason Wang", "email": "jasowang@redhat.com", "time": "Thu Nov 20 17:03:05 2014 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 21 00:26:43 2014 -0500" }, "message": "virtio-net: validate features during probe\n\nWe currently trigger BUG when VIRTIO_NET_F_CTRL_VQ\nis not set but one of features depending on it is.\nThat\u0027s not a friendly way to report errors to\nhypervisors.\nLet\u0027s check, and fail probe instead.\n\nCc: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nCc: Cornelia Huck \u003ccornelia.huck@de.ibm.com\u003e\nCc: Wanlong Gao \u003cgaowanlong@cn.fujitsu.com\u003e\nSigned-off-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nSigned-off-by: Jason Wang \u003cjasowang@redhat.com\u003e\nAcked-by: Cornelia Huck \u003ccornelia.huck@de.ibm.com\u003e\nAcked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7e09dccd07518729fe3cf586beb83acffa2e64ca", "tree": "2ef7558f6db8bd9fd7355d8194d1b5d871edf63f", "parents": [ "c8577819009473311d5f74112e136f17d7859ee1", "f0b4eeced518c632210ef2aea44fc92cc9e86cce" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 21 00:12:39 2014 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 21 00:12:39 2014 -0500" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf\n\nPablo Neira Ayuso says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nNetfilter fixes for net\n\nThe following patchset contains two bugfixes for your net tree, they are:\n\n1) Validate netlink group from nfnetlink to avoid an out of bound array\n access. This should only happen with superuser priviledges though.\n Discovered by Andrey Ryabinin using trinity.\n\n2) Don\u0027t push ethernet header before calling the netfilter output hook\n for multicast traffic, this breaks ebtables since it expects to see\n skb-\u003edata pointing to the network header, patch from Linus Luessing.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c8577819009473311d5f74112e136f17d7859ee1", "tree": "ac3c36ed5d6abf32593a6bb636fbcf0e34af0868", "parents": [ "17544e2ad78fa0bbff6fcdbf09426d04ce95ed1e", "a1d69c60c44134f64945bbf6a6dfda22eaf4a214" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 21 00:07:51 2014 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 21 00:07:51 2014 -0500" }, "message": "Merge tag \u0027master-2014-11-20\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless\n\nJohn W. Linville says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\npull request: wireless 2014-11-20\n\nPlease full this little batch of fixes intended for the 3.18 stream!\n\nFor the mac80211 patch, Johannes says:\n\n\"Here\u0027s another last minute fix, for minstrel HT crashing\ndepending on the value of some uninitialised stack.\"\n\nOn top of that...\n\nBen Greear fixes an ath9k regression in which a BSSID mask is\nmiscalculated.\n\nDmitry Torokhov corrects an error handling routing in brcmfmac which\nwas checking an unsigned variable for a negative value.\n\nJohannes Berg avoids a build problem in brcmfmac for arches where\nlinux/unaligned/access_ok.h and asm/unaligned.h conflict.\n\nMathy Vanhoef addresses another brcmfmac issue so as to eliminate a\nuse-after-free of the URB transfer buffer if a timeout occurs.\n\nPlease let me know if there are problems!\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "17544e2ad78fa0bbff6fcdbf09426d04ce95ed1e", "tree": "8293dc15dfffcb0c689aa25b6f3bbc064f3c20db", "parents": [ "01462405f0c093b2f8dfddafcadcda6c9e4c5cdf" ], "author": { "name": "Anish Bhatt", "email": "anish@chelsio.com", "time": "Thu Nov 20 17:11:46 2014 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 21 00:05:55 2014 -0500" }, "message": "cxgb4 : Fix DCB priority groups being returned in wrong order\n\nPeer priority groups were being reversed, but this was missed in the previous\nfix sent out for this issue.\n\nv2 : Previous patch was doing extra unnecessary work, result is the same.\nPlease ignore previous patch\n\nFixes :\tee7bc3cdc270 (\u0027cxgb4 : dcb open-lldp interop fixes\u0027)\n\nSigned-off-by: Anish Bhatt \u003canish@chelsio.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "01462405f0c093b2f8dfddafcadcda6c9e4c5cdf", "tree": "b405f0cb9c2f653335ef88507f8ba6d16600c87a", "parents": [ "d3052bb5d306b29c1e7d9e5998c5ac4ca1ff0ca9" ], "author": { "name": "Jiri Bohac", "email": "jbohac@suse.cz", "time": "Wed Nov 19 23:05:49 2014 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 20 22:57:03 2014 -0500" }, "message": "ipx: fix locking regression in ipx_sendmsg and ipx_recvmsg\n\nThis fixes an old regression introduced by commit\nb0d0d915 (ipx: remove the BKL).\n\nWhen a recvmsg syscall blocks waiting for new data, no data can be sent on the\nsame socket with sendmsg because ipx_recvmsg() sleeps with the socket locked.\n\nThis breaks mars-nwe (NetWare emulator):\n- the ncpserv process reads the request using recvmsg\n- ncpserv forks and spawns nwconn\n- ncpserv calls a (blocking) recvmsg and waits for new requests\n- nwconn deadlocks in sendmsg on the same socket\n\nCommit b0d0d915 has simply replaced BKL locking with\nlock_sock/release_sock. Unlike now, BKL got unlocked while\nsleeping, so a blocking recvmsg did not block a concurrent\nsendmsg.\n\nOnly keep the socket locked while actually working with the socket data and\nrelease it prior to calling skb_recv_datagram().\n\nSigned-off-by: Jiri Bohac \u003cjbohac@suse.cz\u003e\nReviewed-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d3052bb5d306b29c1e7d9e5998c5ac4ca1ff0ca9", "tree": "16383bb53c7b2edd338cb465ea6bfc41b368fad6", "parents": [ "a5f6fc28d6e6cc379c6839f21820e62262419584" ], "author": { "name": "Joe Stringer", "email": "joestringer@nicira.com", "time": "Wed Nov 19 13:54:49 2014 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 20 22:56:13 2014 -0500" }, "message": "openvswitch: Don\u0027t validate IPv6 label masks.\n\nWhen userspace doesn\u0027t provide a mask, OVS datapath generates a fully\nunwildcarded mask for the flow by copying the flow and setting all bits\nin all fields. For IPv6 label, this creates a mask that matches on the\nupper 12 bits, causing the following error:\n\nopenvswitch: netlink: Invalid IPv6 flow label value (value\u003dffffffff, max\u003dfffff)\n\nThis patch ignores the label validation check for masks, avoiding this\nerror.\n\nSigned-off-by: Joe Stringer \u003cjoestringer@nicira.com\u003e\nAcked-by: Pravin B Shelar \u003cpshelar@nicira.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a5f6fc28d6e6cc379c6839f21820e62262419584", "tree": "f251bb2b01c2465cd791c4db2d5ef2dfe9ef83b6", "parents": [ "ee7255ada313a6db99be47ce174b0bfb8295a041" ], "author": { "name": "Mathias Krause", "email": "minipli@googlemail.com", "time": "Wed Nov 19 18:05:26 2014 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 20 22:43:47 2014 -0500" }, "message": "pptp: fix stack info leak in pptp_getname()\n\npptp_getname() only partially initializes the stack variable sa,\nparticularly only fills the pptp part of the sa_addr union. The code\nthereby discloses 16 bytes of kernel stack memory via getsockname().\n\nFix this by memset(0)\u0027ing the union before.\n\nCc: Dmitry Kozlov \u003cxeb@mail.ru\u003e\nSigned-off-by: Mathias Krause \u003cminipli@googlemail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a0fc608178a9b38a5f782331909fcc208b742a7b", "tree": "1ad7ebdd09ebf9ca8be6edcca2e6541c6a744d1c", "parents": [ "92ff59a696f9917265efa8d369b92e6c9a9a1904", "28731d5818ae25b92d1fb82fe0ac196e97102c1b" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Nov 21 12:19:19 2014 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Nov 21 12:19:19 2014 +1000" }, "message": "Merge branch \u0027drm-fixes-3.18\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-fixes\n\nfix one regression and one endian issue.\n\n* \u0027drm-fixes-3.18\u0027 of git://people.freedesktop.org/~agd5f/linux:\n drm/radeon: fix endian swapping in vbios fetch for tdp table\n drm/radeon: disable native backlight control on pre-r6xx asics (v2)\n" }, { "commit": "b5e212a3051b65e426a513901d9c7001681c7215", "tree": "61ee6c5bb71364d1436136cfb4c1e318f533594b", "parents": [ "70b61e362187b5fccac206506d402f3424e3e749" ], "author": { "name": "Andy Lutomirski", "email": "luto@amacapital.net", "time": "Wed Nov 19 13:56:19 2014 -0800" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Thu Nov 20 23:01:53 2014 +0100" }, "message": "x86, syscall: Fix _TIF_NOHZ handling in syscall_trace_enter_phase1\n\nTIF_NOHZ is 19 (i.e. _TIF_SYSCALL_TRACE | _TIF_NOTIFY_RESUME |\n_TIF_SINGLESTEP), not (1\u003c\u003c19).\n\nThis code is involved in Dave\u0027s trinity lockup, but I don\u0027t see why\nit would cause any of the problems he\u0027s seeing, except inadvertently\nby causing a different path through entry_64.S\u0027s syscall handling.\n\nSigned-off-by: Andy Lutomirski \u003cluto@amacapital.net\u003e\nCc: Don Zickus \u003cdzickus@redhat.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Dave Jones \u003cdavej@redhat.com\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: http://lkml.kernel.org/r/a6cd3b60a3f53afb6e1c8081b0ec30ff19003dd7.1416434075.git.luto@amacapital.net\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "a1d69c60c44134f64945bbf6a6dfda22eaf4a214", "tree": "2678d3082dd3a5822644ca4f24b506d7acc6f343", "parents": [ "6158fb37d1b0448252864b194d585e73e4b94406" ], "author": { "name": "Johannes Berg", "email": "johannes.berg@intel.com", "time": "Wed Nov 19 22:13:10 2014 +0100" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Thu Nov 20 14:46:45 2014 -0500" }, "message": "brcmfmac: don\u0027t include linux/unaligned/access_ok.h\n\nThis is a specific implementation, \u003casm/unaligned.h\u003e is the\nmultiplexer that has the arch-specific knowledge of which\nof the implementations needs to be used, so include that.\n\nThis issue was revealed by kbuild testing\nwhen \u003casm/unaligned.h\u003e was added in \u003clinux/ieee80211.h\u003e\nresulting in redefinition of get_unaligned_be16 (and\nprobably others).\n\nCc: stable@vger.kernel.org # v3.17\nReported-by: Fengguang Wu \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nSigned-off-by: Arend van Spriel \u003carend@broadcom.com\u003e\nSigned-off-by: John W. Linville \u003clinville@tuxdriver.com\u003e\n" }, { "commit": "28731d5818ae25b92d1fb82fe0ac196e97102c1b", "tree": "585d5457dc5613e466e68937173634f1662393db", "parents": [ "b7bc596ebbe0cddc97d76ef9309f64471bbf13eb" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Nov 12 19:17:02 2014 -0500" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Nov 20 11:13:18 2014 -0500" }, "message": "drm/radeon: fix endian swapping in vbios fetch for tdp table\n\nValue needs to be swapped on BE.\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "b7bc596ebbe0cddc97d76ef9309f64471bbf13eb", "tree": "1856c2479af75c1f10b769dba3e7d5491fcaef5d", "parents": [ "bcfef97398972de555e6e1dbf4dac33f0db38d52" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Nov 19 13:12:54 2014 -0500" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Thu Nov 20 11:12:55 2014 -0500" }, "message": "drm/radeon: disable native backlight control on pre-r6xx asics (v2)\n\nJust use the acpi interface. That\u0027s what windows uses on this\ngeneration and it\u0027s the only thing that seems to work reliably\non these generation parts.\n\nYou can still force the native backlight interface by setting\nradeon.backlight\u003d1\n\nBug:\nhttps://bugzilla.kernel.org/show_bug.cgi?id\u003d88501\n\nv2: merge into above if/else block\n\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "7676895f4736421ebafc48de5078e25ea69e88ee", "tree": "19de6c0530dd166d501dda8506f3564b452e7da4", "parents": [ "1d113735ecf21de74a04c3b58fa106ac2e64ca0d" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:02 2014 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:02 2014 +0100" }, "message": "ovl: ovl_dir_fsync() cleanup\n\nCheck against !OVL_PATH_LOWER instead of OVL_PATH_MERGE. For a copied up\ndirectory the two are currently equivalent.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\n" }, { "commit": "1d113735ecf21de74a04c3b58fa106ac2e64ca0d", "tree": "0bb479adc18d97b192719707de0235afac4d0a33", "parents": [ "c9f00fdb9ab3999cb2fb582ad82a5db9e70c82f5" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:01 2014 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:01 2014 +0100" }, "message": "ovl: update MAINTAINERS\n\nThere\u0027s a union/overlay specific mailing list now. Also add a git tree.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\n" }, { "commit": "c9f00fdb9ab3999cb2fb582ad82a5db9e70c82f5", "tree": "49464b103216a446cd75e82ffb6f4c4721fdd33d", "parents": [ "71d509280f7e92eb60ae6b7c78c20afafff060c7" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:01 2014 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:01 2014 +0100" }, "message": "ovl: pass dentry into ovl_dir_read_merged()\n\nPass dentry into ovl_dir_read_merged() insted of upperpath and lowerpath.\nThis cleans up callers and paves the way for multi-layer directory reads.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\n" }, { "commit": "71d509280f7e92eb60ae6b7c78c20afafff060c7", "tree": "c241d261feb9331eaee8dd1137884e86d2c3aae9", "parents": [ "91c77947133f7aef851b625701e182d3f99d14a9" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:01 2014 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:01 2014 +0100" }, "message": "ovl: use lockless_dereference() for upperdentry\n\nDon\u0027t open code lockless_dereference() in ovl_upperdentry_dereference().\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\n" }, { "commit": "91c77947133f7aef851b625701e182d3f99d14a9", "tree": "9ba1e0db29e4b7e6835d32bae45ec45b62371896", "parents": [ "521484639ec19a6f1ed56de6993feb255f5f676c" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:00 2014 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:00 2014 +0100" }, "message": "ovl: allow filenames with comma\n\nAllow option separator (comma) to be escaped with backslash.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\n" }, { "commit": "521484639ec19a6f1ed56de6993feb255f5f676c", "tree": "53cad82bc4ae05b1d2d21a22a2ccde9658e5c8c7", "parents": [ "a105d685a8483985a01776411de191a726b48132" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:00 2014 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:40:00 2014 +0100" }, "message": "ovl: fix race in private xattr checks\n\nXattr operations can race with copy up. This does not matter as long as\nwe consistently fiter out \"trunsted.overlay.opaque\" attribute on upper\ndirectories.\n\nPreviously we checked parent against OVL_PATH_MERGE. This is too general,\nand prone to race with copy-up. I.e. we found the parent to be on the\nlower layer but ovl_dentry_real() would return the copied-up dentry,\npossibly with the \"opaque\" attribute.\n\nSo instead use ovl_path_real() and decide to filter the attributes based on\nthe actual type of the dentry we\u0027ll use.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\n" }, { "commit": "a105d685a8483985a01776411de191a726b48132", "tree": "4296cad5216bb6b71ee417041198db32e7abad14", "parents": [ "ef94b1864d1ed5be54376404bb23d22ed0481feb" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:39:59 2014 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:39:59 2014 +0100" }, "message": "ovl: fix remove/copy-up race\n\novl_remove_and_whiteout() needs to check if upper dentry exists or not\nafter having locked upper parent directory.\n\nPreviously we used a \"type\" value computed before locking the upper parent\ndirectory, which is susceptible to racing with copy-up.\n\nThere\u0027s a similar check in ovl_check_empty_and_clear(). This one is not\nactually racy, since copy-up doesn\u0027t change the \"emptyness\" property of a\ndirectory. Add a comment to this effect, and check the existence of upper\ndentry locally to make the code cleaner.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\n" }, { "commit": "ef94b1864d1ed5be54376404bb23d22ed0481feb", "tree": "32ce8ec582b67b29edac5fd12b5e5679b638dac6", "parents": [ "fc14f9c1272f62c3e8d01300f52467c0d9af50f9" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:39:59 2014 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@suse.cz", "time": "Thu Nov 20 16:39:59 2014 +0100" }, "message": "ovl: rename filesystem type to \"overlay\"\n\nSome distributions carry an \"old\" format of overlayfs while mainline has a\n\"new\" format.\n\nThe distros will possibly want to keep the old overlayfs alongside the new\nfor compatibility reasons.\n\nTo make it possible to differentiate the two versions change the name of\nthe new one from \"overlayfs\" to \"overlay\".\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nReported-by: Serge Hallyn \u003cserge.hallyn@ubuntu.com\u003e\nCc: Andy Whitcroft \u003capw@canonical.com\u003e\n" }, { "commit": "788ec2fc2ca295a2d929986e95231214ecd8d142", "tree": "17f8d173bf4482bbdde8b7ce9c30d5a10ac033f5", "parents": [ "c1a2086e2d8c4eb4e8630ba752e911ec180dec67" ], "author": { "name": "Grant Likely", "email": "grant.likely@linaro.org", "time": "Wed Nov 19 17:13:44 2014 +0000" }, "committer": { "name": "Grant Likely", "email": "grant.likely@linaro.org", "time": "Thu Nov 20 15:32:49 2014 +0000" }, "message": "of/selftest: Fix testing when /aliases is missing\n\nThe /aliases node isn\u0027t always present in the device tree, but the\nunittest code assumes that /aliases is there. Add a check when inserting\nthe testcase data to see if of_aliases needs to be updated, and undo the\nsettings when the nodes are removed.\n\nSigned-off-by: Grant Likely \u003cgrant.likely@linaro.org\u003e\nCc: Rob Herring \u003crobh+dt@kernel.org\u003e\nCc: Gaurav Minocha \u003cgaurav.minocha.os@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\n" }, { "commit": "b1a5ad006b34ded9dc7ec64988deba1b3ecad367", "tree": "04564dc1534581d24bc8152b950572699fd1fc5c", "parents": [ "b6932ee35f1c1364dcea0e691b2feb912a6777e5" ], "author": { "name": "Chris Moore", "email": "Chris.Moore@Emulex.Com", "time": "Tue Nov 04 16:28:29 2014 +0000" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Wed Nov 19 22:28:42 2014 -0800" }, "message": "IB/isert: Adjust CQ size to HW limits\n\nisert has an issue of trying to create a CQ with more CQEs than are\nsupported by the hardware, that currently results in failures during\nisert_device creation during first session login.\n\nThis is the isert version of the patch that Minh Tran submitted for\niser, and is simple a workaround required to function with existing\nocrdma hardware.\n\nSigned-off-by: Chris Moore \u003cchris.moore@emulex.com\u003e\nReviewied-by: Sagi Grimberg \u003csagig@mellanox.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # 3.10+\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "92ff59a696f9917265efa8d369b92e6c9a9a1904", "tree": "0b16b5c79ba847812a7e745cd1f36ff48894e13a", "parents": [ "fc14f9c1272f62c3e8d01300f52467c0d9af50f9", "0485c9dc24ec0939b42ca5104c0373297506b555" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 20 12:58:11 2014 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 20 12:58:11 2014 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2014-11-19\u0027 of git://anongit.freedesktop.org/drm-intel into drm-fixes\n\ntwo regression fixes.\n\n* tag \u0027drm-intel-fixes-2014-11-19\u0027 of git://anongit.freedesktop.org/drm-intel:\n drm/i915: Kick fbdev before vgacon\n drm/i915: drop WaSetupGtModeTdRowDispatch:snb\n" }, { "commit": "78579b7c7eb45f0e7ec5e9437087ed21749f9a9c", "tree": "079b23f3867daa6d02c5f5d3e5081b869fd3ca39", "parents": [ "fc14f9c1272f62c3e8d01300f52467c0d9af50f9" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Nov 19 01:44:11 2014 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Thu Nov 20 01:24:16 2014 +0100" }, "message": "ACPI / PM: Ignore wakeup setting if the ACPI companion can\u0027t wake up\n\nAs reported by Dmitry, on some Chromebooks there are devices with\ncorresponding ACPI objects and with unusual system wakeup\nconfiguration. Namely, they technically are wakeup-capable, but the\nwakeup is handled via a platform-specific out-of-band mechanism and\nthe ACPI PM layer has no information on the wakeup capability. As\na result, device_may_wakeup(dev) called from acpi_dev_suspend_late()\nreturns \u0027true\u0027 for those devices, but the wakeup.flags.valid flag is\nunset for the corresponding ACPI device objects, so acpi_device_wakeup()\nreproducibly fails for them causing acpi_dev_suspend_late() to return\nan error code. The entire system suspend is then aborted and the\nmachines in question cannot suspend at all.\n\nAddress the problem by ignoring the device_may_wakeup(dev) return\nvalue in acpi_dev_suspend_late() if the ACPI companion of the device\nbeing handled has wakeup.flags.valid unset (in which case it is clear\nthat the wakeup is supposed to be handled by other means).\n\nThis fixes a regression introduced by commit a76e9bd89ae7 (i2c:\nattach/detach I2C client device to the ACPI power domain) as the\naffected systems could suspend and resume successfully before that\ncommit.\n\nFixes: a76e9bd89ae7 (i2c: attach/detach I2C client device to the ACPI power domain)\nReported-by: Dmitry Torokhov \u003cdtor@chromium.org\u003e\nReviewed-by: Dmitry Torokhov \u003cdtor@chromium.org\u003e\nCc: 3.13+ \u003cstable@vger.kernel.org\u003e # 3.13+\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n" }, { "commit": "ee7255ada313a6db99be47ce174b0bfb8295a041", "tree": "068f848d1f7921ea58036a0f02935e107db4c1d9", "parents": [ "ffb1388a364d135810337182d6800a0c7ee44f48" ], "author": { "name": "Anish Bhatt", "email": "anish@chelsio.com", "time": "Tue Nov 18 19:09:51 2014 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 19 16:59:26 2014 -0500" }, "message": "cxgb4i : Don\u0027t block unload/cxgb4 unload when remote closes TCP connection\n\ncxgb4i was returning wrong error and not releasing module reference if remote\nend abruptly closed TCP connection. This prevents the cxgb4 network module from\nbeing unloaded, further affecting other network drivers dependent on cxgb4\n\nSending to net as this affects all cxgb4 based network drivers.\n\nSigned-off-by: Anish Bhatt \u003canish@chelsio.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ffb1388a364d135810337182d6800a0c7ee44f48", "tree": "84886bfc3c910d2040274f4cd8cfca470f2bac64", "parents": [ "ddecab1abc97b45168599a52c334a7f9910d5a5a" ], "author": { "name": "Duan Jiong", "email": "duanj.fnst@cn.fujitsu.com", "time": "Wed Nov 19 09:35:39 2014 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 19 16:56:17 2014 -0500" }, "message": "ipv6: delete protocol and unregister rtnetlink when cleanup\n\npim6_protocol was added when initiation, but it not deleted.\nSimilarly, unregister RTNL_FAMILY_IP6MR rtnetlink.\n\nSigned-off-by: Duan Jiong \u003cduanj.fnst@cn.fujitsu.com\u003e\nReviewed-by: Cong Wang \u003ccwang@twopensource.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7fc986d8a9727e5d40da3c2c1c343da6142e82a9", "tree": "0f807067986ae5145945395aa499dd15c2291dac", "parents": [ "7a1562d4f2d01721ad07c3a326db7512077ceea9" ], "author": { "name": "Yinghai Lu", "email": "yinghai@kernel.org", "time": "Wed Nov 19 14:30:32 2014 -0700" }, "committer": { "name": "Bjorn Helgaas", "email": "bhelgaas@google.com", "time": "Wed Nov 19 14:30:32 2014 -0700" }, "message": "PCI: Support 64-bit bridge windows if we have 64-bit dma_addr_t\n\nAaron reported that a 32-bit x86 kernel with Physical Address Extension\n(PAE) support complains about bridge prefetchable memory windows above 4GB:\n\n pci_bus 0000:00: root bus resource [mem 0x380000000000-0x383fffffffff]\n ...\n pci 0000:03:00.0: reg 0x10: [mem 0x383fffc00000-0x383fffdfffff 64bit pref]\n pci 0000:03:00.0: reg 0x20: [mem 0x383fffe04000-0x383fffe07fff 64bit pref]\n pci 0000:03:00.1: reg 0x10: [mem 0x383fffa00000-0x383fffbfffff 64bit pref]\n pci 0000:03:00.1: reg 0x20: [mem 0x383fffe00000-0x383fffe03fff 64bit pref]\n pci 0000:00:02.2: PCI bridge to [bus 03-04]\n pci 0000:00:02.2: bridge window [io 0x1000-0x1fff]\n pci 0000:00:02.2: bridge window [mem 0x91900000-0x91cfffff]\n pci 0000:00:02.2: can\u0027t handle 64-bit address space for bridge\n\nIn this kernel, unsigned long is 32 bits and dma_addr_t is 64 bits.\nPreviously we used \"unsigned long\" to hold the bridge window address. But\nthis is a bus address, so we should use dma_addr_t instead.\n\nUse dma_addr_t to hold the bridge window base and limit.\n\nThe question of whether the CPU can actually *address* the window is\nseparate and depends on what the physical address space of the CPU is and\nwhether the host bridge does any address translation.\n\n[bhelgaas: fix \"shift count \u003e width of type\", changelog, stable tag]\nFixes: d56dbf5bab8c (\"PCI: Allocate 64-bit BARs above 4G when possible\")\nLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d88131\nReported-by: Aaron Ma \u003cmapengyu@gmail.com\u003e\nTested-by: Aaron Ma \u003cmapengyu@gmail.com\u003e\nSigned-off-by: Yinghai Lu \u003cyinghai@kernel.org\u003e\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nCC: stable@vger.kernel.org\t# v3.14+" }, { "commit": "6158fb37d1b0448252864b194d585e73e4b94406", "tree": "6749e22522b92adb2f3a6620258ec6ea5419edd3", "parents": [ "4c69f05eaa428e37890daf88b86a567ce615570b", "280ba51d60be6f4ca3347eaa60783314f38df72e" ], "author": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Wed Nov 19 15:44:40 2014 -0500" }, "committer": { "name": "John W. Linville", "email": "linville@tuxdriver.com", "time": "Wed Nov 19 15:44:40 2014 -0500" }, "message": "Merge tag \u0027mac80211-for-john-2014-11-18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211\n\nJohannes Berg \u003cjohannes@sipsolutions.net\u003e says:\n\n\"Here\u0027s another last minute fix, for minstrel HT crashing\ndepending on the value of some uninitialised stack.\"\n\nSigned-off-by: John W. Linville \u003clinville@tuxdriver.com\u003e\n" }, { "commit": "ddecab1abc97b45168599a52c334a7f9910d5a5a", "tree": "aaacf93b2643afc07f7c62972723555791255c87", "parents": [ "9737c6ab7afbc950e997ef80cba2c40dbbd16ea4", "80646733f11c2e9de3b6339f7e635047e6087280" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 19 15:28:58 2014 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 19 15:28:58 2014 -0500" }, "message": "Merge tag \u0027linux-can-fixes-for-3.18-20141118\u0027 of git://gitorious.org/linux-can/linux-can\n\nMarc Kleine-Budde says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\npull-request: can 2014-11-18\n\nthis is a pull request of 17 patches for net/master for the v3.18 release\ncycle.\n\nThe last patch of this pull request (\"can: m_can: update to support CAN FD\nfeatures\") adds, as the description says, a new feature to the m_can driver. As\nthe m_can driver has been added in v3.18 there is no risk of causing a\nregression. Give me a note if this is not okay and I\u0027ll create a new pull\nrequest without it.\n\nThere is a patch for the CAN infrastructure by Thomas Körper which fixes\ncalling kfree_skb() from interrupt context. Roman Fietze fixes a typo also in\nthe infrastructure. A patch by Dong Aisheng adds a generic helper function to\ntell if a skb is normal CAN or CAN-FD frame. Alexey Khoroshilov of the Linux\nDriver Verification project fixes a memory leak in the esd_usb2 driver. Two\npatches by Sudip Mukherjee remove unused variables and fixe the signess of a\nvariable. Three patches by me add the missing .ndo_change_mtu callback to the\nxilinx_can, rcar_can and gs_usb driver.\n\nThe remaining patches improve the m_can driver: David Cohen adds the missing\nCONFIG_HAS_IOMEM dependency. Dong Aisheng provides 6 bugfix patches (most\nimportant: missing RAM init, sleep in NAPI poll, dlc in RTR). While the last of\nhis patches adds CAN FD support to the driver.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9737c6ab7afbc950e997ef80cba2c40dbbd16ea4", "tree": "67758ef137d3fd1c61885229d1e3221538d30686", "parents": [ "b8e4500f42fe4464a33a887579147050bed8fcef" ], "author": { "name": "Or Gerlitz", "email": "ogerlitz@mellanox.com", "time": "Tue Nov 18 17:51:27 2014 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 19 15:11:09 2014 -0500" }, "message": "net/mlx4_en: Add VXLAN ndo calls to the PF net device ops too\n\nThis is currently missing, which results in a crash when one attempts\nto set VXLAN tunnel over the mlx4_en when acting as PF.\n\n\t[ 2408.785472] BUG: unable to handle kernel NULL pointer dereference at (null)\n\t[...]\n\t[ 2408.994104] Call Trace:\n\t[ 2408.996584] [\u003cffffffffa021f7f5\u003e] ? vxlan_get_rx_port+0xd6/0x103 [vxlan]\n\t[ 2409.003316] [\u003cffffffffa021f71f\u003e] ? vxlan_lowerdev_event+0xf2/0xf2 [vxlan]\n\t[ 2409.010225] [\u003cffffffffa0630358\u003e] mlx4_en_start_port+0x862/0x96a [mlx4_en]\n\t[ 2409.017132] [\u003cffffffffa063070f\u003e] mlx4_en_open+0x17f/0x1b8 [mlx4_en]\n\nWhile here, make sure to invoke vxlan_get_rx_port() only when VXLAN\noffloads are actually enabled and not when they are only supported.\n\nReported-by: Ido Shamay \u003cidos@mellanox.com\u003e\nSigned-off-by: Or Gerlitz \u003cogerlitz@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b8e4500f42fe4464a33a887579147050bed8fcef", "tree": "839cb0a74f0b42777fa848ab5b0d405ccb13ff31", "parents": [ "11bf7828a59880427403e13dcff8228d67e9e0f7" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@redhat.com", "time": "Tue Nov 18 15:14:44 2014 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 19 15:09:04 2014 -0500" }, "message": "bonding: fix curr_active_slave/carrier with loadbalance arp monitoring\n\nSince commit 6fde8f037e60 (\"bonding: fix locking in\nbond_loadbalance_arp_mon()\") we can have a stale bond carrier state and\nstale curr_active_slave when using arp monitoring in loadbalance modes. The\nreason is that in bond_loadbalance_arp_mon() we can\u0027t have\ndo_failover \u003d\u003d true but slave_state_changed \u003d\u003d false, whenever do_failover\nis true then slave_state_changed is also true. Then the following piece\nfrom bond_loadbalance_arp_mon():\n if (slave_state_changed) {\n bond_slave_state_change(bond);\n if (BOND_MODE(bond) \u003d\u003d BOND_MODE_XOR)\n bond_update_slave_arr(bond, NULL);\n } else if (do_failover) {\n block_netpoll_tx();\n bond_select_active_slave(bond);\n unblock_netpoll_tx();\n }\n\nwill execute only the first branch, always and regardless of do_failover.\nSince these two events aren\u0027t related in such way, we need to decouple and\nconsider them separately.\n\nFor example this issue could lead to the following result:\nBonding Mode: load balancing (round-robin)\n*MII Status: down*\nMII Polling Interval (ms): 0\nUp Delay (ms): 0\nDown Delay (ms): 0\nARP Polling Interval (ms): 100\nARP IP target/s (n.n.n.n form): 192.168.9.2\n\nSlave Interface: ens12\n*MII Status: up*\nSpeed: 10000 Mbps\nDuplex: full\nLink Failure Count: 2\nPermanent HW addr: 00:0f:53:01:42:2c\nSlave queue ID: 0\n\nSlave Interface: eth1\n*MII Status: up*\nSpeed: Unknown\nDuplex: Unknown\nLink Failure Count: 70\nPermanent HW addr: 52:54:00:2f:0f:8e\nSlave queue ID: 0\n\nSince some interfaces are up, then the status of the bond should also be\nup, but it will never change unless something invokes bond_set_carrier()\n(i.e. enslave, bond_select_active_slave etc). Now, if I force the\ncalling of bond_select_active_slave via for example changing\nprimary_reselect (it can change in any mode), then the MII status goes to\n\"up\" because it calls bond_select_active_slave() which should\u0027ve been done\nfrom bond_loadbalance_arp_mon() itself.\n\nCC: Veaceslav Falico \u003cvfalico@gmail.com\u003e\nCC: Jay Vosburgh \u003cj.vosburgh@gmail.com\u003e\nCC: Andy Gospodarek \u003candy@greyhouse.net\u003e\nCC: Ding Tianhong \u003cdingtianhong@huawei.com\u003e\n\nFixes: 6fde8f037e60 (\"bonding: fix locking in bond_loadbalance_arp_mon()\")\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@redhat.com\u003e\nAcked-by: Veaceslav Falico \u003cvfalico@gmail.com\u003e\nAcked-by: Andy Gospodarek \u003cgospo@cumulusnetworks.com\u003e\nAcked-by: Ding Tianhong \u003cdingtianhong@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f82c458a2c3ffb94b431fc6ad791a79df1b3713e", "tree": "0eca3f95f74d1cde140366002b3cd794ce96f67c", "parents": [ "6e5aafb27419f32575b27ef9d6a31e5d54661aca" ], "author": { "name": "Chris Mason", "email": "clm@fb.com", "time": "Wed Nov 19 10:25:09 2014 -0800" }, "committer": { "name": "Chris Mason", "email": "clm@fb.com", "time": "Wed Nov 19 10:34:35 2014 -0800" }, "message": "btrfs: fix lockups from btrfs_clear_path_blocking\n\nThe fair reader/writer locks mean that btrfs_clear_path_blocking needs\nto strictly follow lock ordering rules even when we already have\nblocking locks on a given path.\n\nBefore we can clear a blocking lock on the path, we need to make sure\nall of the locks have been converted to blocking. This will remove lock\ninversions against anyone spinning in write_lock() against the buffers\nwe\u0027re trying to get read locks on. These inversions didn\u0027t exist before\nthe fair read/writer locks, but now we need to be more careful.\n\nWe papered over this deadlock in the past by changing\nbtrfs_try_read_lock() to be a true trylock against both the spinlock and\nthe blocking lock. This was slower, and not sufficient to fix all the\ndeadlocks. This patch adds a btrfs_tree_read_lock_atomic(), which\nbasically means get the spinlock but trylock on the blocking lock.\n\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\nSigned-off-by: Josef Bacik \u003cjbacik@fb.com\u003e\nReported-by: Patrick Schmid \u003cschmid@phys.ethz.ch\u003e\ncc: stable@vger.kernel.org #v3.15+\n" }, { "commit": "7ca2f234404e738cc807ed87c43f9932513bc8c6", "tree": "17b9a0eeb1230a42ae67e11f361aff37b652804b", "parents": [ "4a7795d35e252f38298980530e01e21867f8f856" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 15:11:24 2014 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Nov 19 13:09:37 2014 -0500" }, "message": "isofs: avoid unused function warning\n\nWith the isofs_hash() function removed, isofs_hash_ms() is the only user\nof isofs_hash_common(), but it\u0027s defined inside of an #ifdef, which triggers\nthis gcc warning in ARM axm55xx_defconfig starting with v3.18-rc3:\n\nfs/isofs/inode.c:177:1: warning: \u0027isofs_hash_common\u0027 defined but not used [-Wunused-function]\n\nThis patch moves the function inside of the same #ifdef section to avoid that\nwarning, which seems the best compromise of a relatively harmless patch for\na late -rc.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nFixes: b0afd8e5db7b (\"isofs: don\u0027t bother with -\u003ed_op for normal case\")\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "4a7795d35e252f38298980530e01e21867f8f856", "tree": "1ed969cdba05031ff6410a14f32225744b351022", "parents": [ "ad0eab9293485d1c06237e9249f6d4dfa3d93d4d" ], "author": { "name": "Yan, Zheng", "email": "zyan@redhat.com", "time": "Wed Nov 19 15:50:34 2014 +0800" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Wed Nov 19 13:07:20 2014 -0500" }, "message": "vfs: fix reference leak in d_prune_aliases()\n\nIn \"d_prune_alias(): just lock the parent and call __dentry_kill()\" the old\ndget + d_drop + dput has been replaced with lock_parent + __dentry_kill;\nunfortunately, dput() does more than just killing dentry - it also drops the\nreference to parent. New variant leaks that reference and needs dput(parent)\nafter killing the child off.\n\nSigned-off-by: Yan, Zheng \u003czyan@redhat.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "c1a2086e2d8c4eb4e8630ba752e911ec180dec67", "tree": "a7234d01a8647f98e644e89a3b2675b2de4e1c30", "parents": [ "a0e27f51ba8a04125c22a95c4d3e98297a7191de" ], "author": { "name": "Grant Likely", "email": "grant.likely@linaro.org", "time": "Wed Nov 19 16:22:32 2014 +0000" }, "committer": { "name": "Grant Likely", "email": "grant.likely@linaro.org", "time": "Wed Nov 19 17:42:57 2014 +0000" }, "message": "of/selftest: Fix off-by-one error in removal path\n\nThe removal path for selftest data has an off by one error that causes\nthe code to dereference beyond the end of the nodes[] array on the first\npass through. The old code only worked by chance on a lot of platforms,\nbut the bug was recently exposed on aarch64.\n\nThe fix is simple. Decrement the node count before dereferencing, not\nafter.\n\nReported-by: Kevin Hilman \u003ckhilman@linaro.org\u003e\nCc: Rob Herring \u003crobh+dt@kernel.org\u003e\nCc: Gaurav Minocha \u003cgaurav.minocha.os@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # v3.17+\n" }, { "commit": "49e41938f8a53d01d1c5d133a038acb9650008f9", "tree": "e674598dec5201481255322593c4b062a3ca933f", "parents": [ "4a8ab7713bcc8504228784b7796ae6d5089acd66" ], "author": { "name": "Tyler Baker", "email": "tyler.baker@linaro.org", "time": "Tue Nov 18 11:10:53 2014 -0800" }, "committer": { "name": "Kevin Hilman", "email": "khilman@linaro.org", "time": "Wed Nov 19 09:35:17 2014 -0800" }, "message": "ARM: multi_v7_defconfig: fix failure setting CPU voltage by enabling dependent I2C controller\n\nThis patch fixes a long standing issue introduced during the 3.16 merge window.\nShortly after the merge, exynos5250-based arndale boards began to produce the\nfollowing errors:\n\nkern.err kernel: exynos-cpufreq exynos-cpufreq: failed to set cpu voltage\nkern.err kernel: cpufreq: __target_index: Failed to change cpu frequency: -22\n\nFurther analysis revealed that the S5M8767 voltage regulator used on the\nexynos5250-based arndale board utilizes the S3C2410 I2C controller. If the\nS3C2410 I2C controller driver is not enabled, the S5M8767 voltage regulator\nfails to probe. Therefore a dependency exists between these two drivers.\nIn the exynos_defconfig both CONFIG_REGULATOR_S5M8767 and CONFIG_I2C_S3C2410\noptions are enabled, and no errors are produced. However, in the\nmulti_v7_defconfig only the CONFIG_REGULATOR_S5M8767 option is enabled and the\nerrors are present. So let\u0027s enable the CONFIG_I2C_S3C2410 option in the\nmulti_v7_defconfig to allow the S5M8767 voltage regulator to probe.\n\nSigned-off-by: Tyler Baker \u003ctyler.baker@linaro.org\u003e\nAcked-by: Kukjin Kim \u003ckgene.kim@samsung.com\u003e\nSigned-off-by: Kevin Hilman \u003ckhilman@linaro.org\u003e\n" }, { "commit": "935c2dbec4d6d3163ee8e7409996904a734ad89a", "tree": "e786a6f69c5c5f2f03ac85e5466906a6524fec18", "parents": [ "99436f7d69045800ffd1d66912f85d37150c7e2b" ], "author": { "name": "Maciej W. Rozycki", "email": "macro@codesourcery.com", "time": "Mon Nov 17 16:10:32 2014 +0000" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:09 2014 +0100" }, "message": "MIPS: jump_label.c: Handle the microMIPS J instruction encoding\n\nImplement the microMIPS encoding of the J instruction for the purpose of\nthe static keys feature, fixing a crash early on in bootstrap as the\nkernel is unhappy seeing the ISA bit set in jump table entries. Make\nsure the ISA bit correctly reflects the instruction encoding chosen for\nthe kernel, 0 for the standard MIPS and 1 for the microMIPS encoding.\n\nAlso make sure the instruction to patch is a 32-bit NOP in the microMIPS\nmode as by default the 16-bit short encoding is assumed\n\nSigned-off-by: Maciej W. Rozycki \u003cmacro@codesourcery.com\u003e\nCc: linux-mips@linux-mips.org\nPatchwork: https://patchwork.linux-mips.org/patch/8516/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "99436f7d69045800ffd1d66912f85d37150c7e2b", "tree": "962a10905af43e16423a1a0878fa6a87deae78ea", "parents": [ "640465bda58c7078725201be7430c31a349121e9" ], "author": { "name": "Maciej W. Rozycki", "email": "macro@codesourcery.com", "time": "Mon Nov 17 16:09:54 2014 +0000" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:08 2014 +0100" }, "message": "MIPS: jump_label.c: Correct the span of the J instruction\n\nCorrect the check for the span of the 256MB segment addressable by the J\ninstruction according to this instruction\u0027s semantics. The calculation\nof the jump target is applied to the address of the delay-slot\ninstruction that immediately follows. Adjust the check accordingly by\nadding 4 to `e-\u003ecode\u0027 that holds the address of the J instruction\nitself.\n\nSigned-off-by: Maciej W. Rozycki \u003cmacro@codesourcery.com\u003e\nCc: linux-mips@linux-mips.org\nPatchwork: https://patchwork.linux-mips.org/patch/8515/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "640465bda58c7078725201be7430c31a349121e9", "tree": "d3e9804280f58cbe7638358c519c238b50325e6e", "parents": [ "51b1029d9966060c6ad02030e6f251425b4f06c1" ], "author": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Tue Nov 18 18:47:13 2014 +0100" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:08 2014 +0100" }, "message": "MIPS: Zero variable read by get_user / __get_user in case of an error.\n\nThis wasn\u0027t happening in all cases.\n\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "51b1029d9966060c6ad02030e6f251425b4f06c1", "tree": "727f8e0d2a9fddc98072151c052686df6498e7ed", "parents": [ "6a8dff6ab16c903b0d8ef5fbf21543f39bf5d675" ], "author": { "name": "Markos Chandras", "email": "markos.chandras@imgtec.com", "time": "Mon Nov 17 09:32:38 2014 +0000" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:08 2014 +0100" }, "message": "MIPS: lib: memcpy: Restore NOP on delay slot before returning to caller\n\nCommit cf62a8b8134dd3 (\"MIPS: lib: memcpy: Use macro to build the\ncopy_user code\") switched to a macro in order to build the memcpy\nsymbols in preparation for the EVA support. However, this commit\nalso removed the NOP instruction after the \u0027jr ra\u0027 when returning\nback to the caller. This had no visible side-effects since the next\ninstruction was a load to the t0 register which was already in the\nclobbered list, but it may have undesired effects in the future\nif some other code is introduced in between the .Ldone and\nthe .Ll_exc_copy labels.\n\nSigned-off-by: Markos Chandras \u003cmarkos.chandras@imgtec.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # v3.15+\nCc: linux-mips@linux-mips.org\nPatchwork: https://patchwork.linux-mips.org/patch/8512/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "6a8dff6ab16c903b0d8ef5fbf21543f39bf5d675", "tree": "3faa8cc35a0550e326af511bce32ae69013b4830", "parents": [ "58563817cfed0432e9a54476d5fc6c3aeba475e4" ], "author": { "name": "Markos Chandras", "email": "markos.chandras@imgtec.com", "time": "Mon Nov 17 09:31:07 2014 +0000" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:08 2014 +0100" }, "message": "MIPS: tlb-r4k: Add missing HTW stop/start sequences\n\nHTW needs to stop and start again whenever the EntryHI register\nchanges otherwise an inflight HTW operation might use the new\nEntryHI register for updating an old entry and that could lead\nto crashes or even a machine check exception. We fix this by\nensuring the HTW has stop whenever the EntryHI register is about\nto change\n\nSigned-off-by: Markos Chandras \u003cmarkos.chandras@imgtec.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # v3.17+\nCc: linux-mips@linux-mips.org\nPatchwork: https://patchwork.linux-mips.org/patch/8511/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "58563817cfed0432e9a54476d5fc6c3aeba475e4", "tree": "3c0becc2bde0f5b98b7bbbe482d615247dfafa8b", "parents": [ "bbaf113a481b6ce32444c125807ad3618643ce57" ], "author": { "name": "Markos Chandras", "email": "markos.chandras@imgtec.com", "time": "Mon Nov 17 09:30:23 2014 +0000" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:08 2014 +0100" }, "message": "MIPS: asm: uaccess: Add v1 register to clobber list on EVA\n\nWhen EVA is turned on and prefetching is being used in memcpy.S,\nthe v1 register is being used as a helper register to the PREFE\ninstruction. However, v1 ($3) was not in the clobber list, which\nmeans that the compiler did not preserve it across function calls,\nand that could corrupt the value of the register leading to all\nsorts of userland crashes. We fix this problem by using the\nDADDI_SCRATCH macro to define the clobbered register when\nCONFIG_EVA \u0026\u0026 CONFIG_CPU_HAS_PREFETCH are enabled.\n\nSigned-off-by: Markos Chandras \u003cmarkos.chandras@imgtec.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # v3.15+\nCc: linux-mips@linux-mips.org\nPatchwork: https://patchwork.linux-mips.org/patch/8510/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "bbaf113a481b6ce32444c125807ad3618643ce57", "tree": "50d109a6d7180f9c83caa9ad13678e8a1147d423", "parents": [ "7352c8b13dd9a848b0c5d6209d62761afb164dcb" ], "author": { "name": "Aaro Koskinen", "email": "aaro.koskinen@nsn.com", "time": "Fri Oct 17 18:10:24 2014 +0300" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:07 2014 +0100" }, "message": "MIPS: oprofile: Fix backtrace on 64-bit kernel\n\nFix incorrect cast that always results in wrong address for the new\nframe on 64-bit kernels.\n\nSigned-off-by: Aaro Koskinen \u003caaro.koskinen@nsn.com\u003e\nCc: stable@vger.kernel.org\nCc: linux-mips@linux-mips.org\nPatchwork: https://patchwork.linux-mips.org/patch/8110/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "7352c8b13dd9a848b0c5d6209d62761afb164dcb", "tree": "307be3b53c379ccbcf63815f4f0701735dbac33b", "parents": [ "cc94ea31150f83d1f70ad854c920dd6b739c1628" ], "author": { "name": "Huacai Chen", "email": "chenhc@lemote.com", "time": "Tue Nov 04 14:13:23 2014 +0800" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:07 2014 +0100" }, "message": "MIPS: Loongson: Set Loongson-3\u0027s ISA level to MIPS64R1\n\nIn CPU manual Loongson-3 is MIPS64R2 compatible, but during tests we\nfound that its EI/DI instructions have problems. So we just set the ISA\nlevel to MIPS64R1.\n\nSigned-off-by: Huacai Chen \u003cchenhc@lemote.com\u003e\nCc: John Crispin \u003cjohn@phrozen.org\u003e\nCc: Steven J. Hill \u003cSteven.Hill@imgtec.com\u003e\nCc: linux-mips@linux-mips.org\nCc: Fuxin Zhang \u003czhangfx@lemote.com\u003e\nCc: Zhangjin Wu \u003cwuzhangjin@gmail.com\u003e\nPatchwork: https://patchwork.linux-mips.org/patch/8320/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "cc94ea31150f83d1f70ad854c920dd6b739c1628", "tree": "6d3a36871ffac477f338e148432958029ad8af54", "parents": [ "5829b0ecc584d15ae4eeabe69f2ab554bdec4689" ], "author": { "name": "Huacai Chen", "email": "chenhc@lemote.com", "time": "Tue Nov 04 14:13:22 2014 +0800" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:07 2014 +0100" }, "message": "MIPS: Loongson: Fix the write-combine CCA value setting\n\nAll Loongson-2/3 processors support _CACHE_UNCACHED_ACCELERATED, not\nonly Loongson-3A.\n\nSigned-off-by: Huacai Chen \u003cchenhc@lemote.com\u003e\nCc: John Crispin \u003cjohn@phrozen.org\u003e\nCc: Steven J. Hill \u003cSteven.Hill@imgtec.com\u003e\nCc: linux-mips@linux-mips.org\nCc: Fuxin Zhang \u003czhangfx@lemote.com\u003e\nCc: Zhangjin Wu \u003cwuzhangjin@gmail.com\u003e\nPatchwork: https://patchwork.linux-mips.org/patch/8319/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "5829b0ecc584d15ae4eeabe69f2ab554bdec4689", "tree": "0caee22b32b4ebd68cf36bfcd81682e419338e14", "parents": [ "21255dad9dffa4407cab866f5561cb9568f7f7d8" ], "author": { "name": "James Cowgill", "email": "James.Cowgill@imgtec.com", "time": "Thu Nov 13 11:08:07 2014 +0000" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:07 2014 +0100" }, "message": "MIPS: IP27: Fix __node_distances undefined error\n\nexport the __node_distances symbol in the ip27 memory code to fix the\nbuild error:\n\n Building modules, stage 2.\n MODPOST 311 modules\nERROR: \"__node_distances\" [drivers/block/nvme.ko] undefined!\nscripts/Makefile.modpost:90: recipe for target \u0027__modpost\u0027 failed\n\nwhen building the kernel with:\n CONFIG_SGI_IP27\u003dy\n CONFIG_BLK_DEV_NVME\u003dm\n\nSigned-off-by: James Cowgill \u003cJames.Cowgill@imgtec.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # v3.15+\nReviewed-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "21255dad9dffa4407cab866f5561cb9568f7f7d8", "tree": "5916a1d256ef3b73a4165b0e36740c5e5c65aea3", "parents": [ "206c5f60a3d902bc4b56dab2de3e88de5eb06108" ], "author": { "name": "James Cowgill", "email": "James.Cowgill@imgtec.com", "time": "Thu Nov 13 11:08:06 2014 +0000" }, "committer": { "name": "Ralf Baechle", "email": "ralf@linux-mips.org", "time": "Wed Nov 19 18:22:07 2014 +0100" }, "message": "MIPS: Loongson3: Fix __node_distances undefined error\n\nexport the __node_distances symbol in the loongson3 numa code to fix the\nbuild error:\n\n Building modules, stage 2.\n MODPOST 221 modules\nERROR: \"__node_distances\" [drivers/block/nvme.ko] undefined!\nscripts/Makefile.modpost:90: recipe for target \u0027__modpost\u0027 failed\n\nwhen building the kernel with:\n CONFIG_CPU_LOONGSON3\u003dy\n CONFIG_NUMA\u003dy\n CONFIG_BLK_DEV_NVME\u003dm\n\nSigned-off-by: James Cowgill \u003cJames.Cowgill@imgtec.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # v3.17+\nReviewed-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nReviewed-by: Huacai Chen \u003cchenhc@lemote.com\u003e\nCc: linux-mips@linux-mips.org\nCc: Wei Yongjun \u003cyongjun_wei@trendmicro.com.cn\u003e\nPatchwork: https://patchwork.linux-mips.org/patch/8444/\nSigned-off-by: Ralf Baechle \u003cralf@linux-mips.org\u003e\n" }, { "commit": "4a8ab7713bcc8504228784b7796ae6d5089acd66", "tree": "9e90191791d54e1e8511f3dfef817af798e6f777", "parents": [ "5210436b8180bf406cdc9c85ed09a46bccfd6e64", "221b9bf42b26a22e6904d20f35c53aec2e73a646" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 17:35:30 2014 +0100" }, "committer": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 17:35:30 2014 +0100" }, "message": "Merge tag \u0027tegra-for-3.18-fixes-for-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux into fixes\n\nPull \"ARM: tegra: Device tree fixes for v3.18-rc5\" from Thierry Reding:\n\nThis contains the serial port numbering fixes that are required for the\nserial port numbering to stay the same with or without the serial core\nmaking use of the aliases defined in DT.\n\neMMC is also fixed for TN7 and Roth boards which were using the wrong\nregulators.\n\n* tag \u0027tegra-for-3.18-fixes-for-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:\n ARM: tegra: roth: Fix SD card VDD_IO regulator\n ARM: tegra: Remove eMMC vmmc property for roth/tn7\n ARM: dts: tegra: move serial aliases to per-board\n ARM: tegra: Add serial port labels to Tegra124 DT\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n" }, { "commit": "5210436b8180bf406cdc9c85ed09a46bccfd6e64", "tree": "7ec7204ee373a8b548e479f774cb22396cd45da8", "parents": [ "3410d4247cdbadfd08b455adf9217404e0eb71ba", "4f37828d4d69a46830e0525a065da9847fc7a819" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 17:26:52 2014 +0100" }, "committer": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 17:26:52 2014 +0100" }, "message": "Merge tag \u0027renesas-clock-fixes-for-v3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into fixes\n\nPull \"Renesas ARM Based SoC Clock Fixes for v3.18\" from Simon Horman:\n\n* Correct IIC0 parent clock for r8a7740\n* Add missing INTCA clock for irqpin module for r8a7740\n* Correct SD3CKCR address on r8a7790\n\n* tag \u0027renesas-clock-fixes-for-v3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:\n ARM: shmobile: r8a7740 legacy: Correct IIC0 parent clock\n ARM: shmobile: r8a7740 legacy: Add missing INTCA clock for irqpin module\n ARM: shmobile: r8a7790: Fix SD3CKCR address\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n" }, { "commit": "3410d4247cdbadfd08b455adf9217404e0eb71ba", "tree": "e7cf62a2e48afef76be7d3000ec514b39a386025", "parents": [ "1b6166e5bac2d36ecf83ded7e5c863b087862441", "b89ff7c3c2dee189489a5f45eb8d72e106179299" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 17:25:59 2014 +0100" }, "committer": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 17:25:59 2014 +0100" }, "message": "Merge tag \u0027renesas-dt-fixes-for-v3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into fixes\n\nPull \"Renesas ARM Based SoC DT Fixes for v3.18\" from Simon Horman:\n\n* Correct IIC0 parent clock on r8a7740\n* Correct SD3CKCR address to device tree on r8a7790\n\n* tag \u0027renesas-dt-fixes-for-v3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:\n ARM: shmobile: r8a7740 dtsi: Correct IIC0 parent clock\n ARM: shmobile: r8a7790: Fix SD3CKCR address to device tree\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n" }, { "commit": "1b6166e5bac2d36ecf83ded7e5c863b087862441", "tree": "379c1df0fa1bde4340bf32dab38d8cf2ec304020", "parents": [ "4e2594c4df50d7b8d9fd3defcf81082ef3b45638", "b207422bb92f17f4f892a9b8737d44b37fece25b" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 17:22:12 2014 +0100" }, "committer": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 17:22:12 2014 +0100" }, "message": "Merge tag \u0027renesas-soc-fixes-for-v3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into fixes\n\nPull \"Renesas ARM Based SoC Fixes for v3.18\" from Simon Horman:\n\n* Set i2c clks_per_count to 2 on kzm9g\n\n* tag \u0027renesas-soc-fixes-for-v3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:\n ARM: shmobile: kzm9g legacy: Set i2c clks_per_count to 2\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n" }, { "commit": "4e2594c4df50d7b8d9fd3defcf81082ef3b45638", "tree": "13fafb57a8b8b21f7cad245ddb0a90469bec81cc", "parents": [ "fc14f9c1272f62c3e8d01300f52467c0d9af50f9", "532425a7a7738b479406496ca4ad6b3247688e44" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 14:16:16 2014 +0100" }, "committer": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Wed Nov 19 14:16:16 2014 +0100" }, "message": "Merge tag \u0027sunxi-fixes-for-3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mripard/linux into fixes\n\nMerge \"Allwinner fixes for 3.18\" from Maxime Ripard:\n\nA fix for the A31 dma controller that requires the AHB clock to be parented to\nPLL6 in order to operate.\n\n* tag \u0027sunxi-fixes-for-3.18\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mripard/linux:\n ARM: dts: sun6i: Re-parent ahb1_mux to pll6 as required by dma controller\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n" }, { "commit": "6bab4a8a1888729f17f4923cc5867e4674f66333", "tree": "95cf8a083e1799d3fabbf354ed40ac4bc599d19b", "parents": [ "59aa896db80479dec29f471a7ca2b9eeeeb7d38e" ], "author": { "name": "Maxime Ripard", "email": "maxime.ripard@free-electrons.com", "time": "Tue Nov 18 23:59:33 2014 +0100" }, "committer": { "name": "Daniel Lezcano", "email": "daniel.lezcano@linaro.org", "time": "Wed Nov 19 10:43:51 2014 +0100" }, "message": "clockevent: sun4i: Fix race condition in the probe code\n\nThe interrupts were activated and the handler registered before the clockevent\nwas registered in the probe function.\n\nThe interrupt handler, however, was making the assumption that the clockevent\ndevice was registered.\n\nThat could cause a null pointer dereference if the timer interrupt was firing\nduring this narrow window.\n\nFix that by moving the clockevent registration before the interrupt is enabled.\n\nReported-by: Roman Byshko \u003crbyshko@gmail.com\u003e\nSigned-off-by: Maxime Ripard \u003cmaxime.ripard@free-electrons.com\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n" }, { "commit": "280ba51d60be6f4ca3347eaa60783314f38df72e", "tree": "4b6f66ce8187b59e755f02f14ffd5d488b6084df", "parents": [ "4f031fa9f188b2b0641ac20087d9e16bcfb4e49d" ], "author": { "name": "Felix Fietkau", "email": "nbd@openwrt.org", "time": "Tue Nov 18 22:35:31 2014 +0100" }, "committer": { "name": "Johannes Berg", "email": "johannes.berg@intel.com", "time": "Tue Nov 18 22:39:16 2014 +0100" }, "message": "mac80211: minstrel_ht: fix a crash in rate sorting\n\nThe commit 5935839ad73583781b8bbe8d91412f6826e218a4\n\"mac80211: improve minstrel_ht rate sorting by throughput \u0026 probability\"\n\nintroduced a crash on rate sorting that occurs when the rate added to\nthe sorting array is faster than all the previous rates. Due to an\noff-by-one error, it reads the rate index from tp_list[-1], which\ncontains uninitialized stack garbage, and then uses the resulting index\nfor accessing the group rate stats, leading to a crash if the garbage\nvalue is big enough.\n\nCc: Thomas Huehn \u003cthomas@net.t-labs.tu-berlin.de\u003e\nReported-by: Jouni Malinen \u003cj@w1.fi\u003e\nSigned-off-by: Felix Fietkau \u003cnbd@openwrt.org\u003e\nSigned-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n" }, { "commit": "11bf7828a59880427403e13dcff8228d67e9e0f7", "tree": "35bc60b412b19009bd5bf604b2e8d98f3116d125", "parents": [ "feb91a02ccb09661507f170b2a444aec94f307f9" ], "author": { "name": "Joe Stringer", "email": "joestringer@nicira.com", "time": "Mon Nov 17 16:24:54 2014 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 18 15:38:44 2014 -0500" }, "message": "vxlan: Inline vxlan_gso_check().\n\nSuggested-by: Or Gerlitz \u003cogerlitz@mellanox.com\u003e\nSigned-off-by: Joe Stringer \u003cjoestringer@nicira.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "80646733f11c2e9de3b6339f7e635047e6087280", "tree": "66278a4176be077e2478ac72c5d2ea5b9154454b", "parents": [ "a93f5cae67b366e4ce7e9eb336e2885309fd6612" ], "author": { "name": "Dong Aisheng", "email": "b29396@freescale.com", "time": "Tue Nov 18 19:00:55 2014 +0800" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:35:06 2014 +0100" }, "message": "can: m_can: update to support CAN FD features\n\nBosch M_CAN is CAN FD capable device. This patch implements the CAN\nFD features include up to 64 bytes payload and bitrate switch function.\n1) Change the Rx FIFO and Tx Buffer to 64 bytes for support CAN FD\n up to 64 bytes payload. It\u0027s backward compatible with old 8 bytes\n normal CAN frame.\n2) Allocate can frame or canfd frame based on EDL bit\n3) Bitrate Switch function is disabled by default and will be enabled\n according to CANFD_BRS bit in cf-\u003eflags.\n\nAcked-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\nSigned-off-by: Dong Aisheng \u003cb29396@freescale.com\u003e\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "a93f5cae67b366e4ce7e9eb336e2885309fd6612", "tree": "e8cdbdc6d742c6529fd5131cfadb3e5913eb8cce", "parents": [ "7660f633070986ab4ee41c063a90e140d5781e13" ], "author": { "name": "Dong Aisheng", "email": "b29396@freescale.com", "time": "Wed Oct 29 18:45:22 2014 +0800" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:35:05 2014 +0100" }, "message": "can: m_can: fix incorrect error messages\n\nFix a few error messages.\n\nSigned-off-by: Dong Aisheng \u003cb29396@freescale.com\u003e\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "7660f633070986ab4ee41c063a90e140d5781e13", "tree": "f28d7b59ee88fac428a96e2e321727de42b2185b", "parents": [ "921f168109d029e3b59459c3a2754f0e2a70fad3" ], "author": { "name": "Dong Aisheng", "email": "b29396@freescale.com", "time": "Wed Oct 29 18:45:24 2014 +0800" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:35:05 2014 +0100" }, "message": "can: m_can: add missing delay after setting CCCR_INIT bit\n\nThe spec mentions there may be a delay until the value written to INIT can be\nread back due to the synchronization mechanism between the two clock domains.\nBut it does not indicate the exact clock cycles needed. The 5us delay is a\ntest value and seems ok.\n\nWithout the delay, CCCR.CCE bit may fail to be set and then the initialization\nfail sometimes when do repeatly up and down.\n\nSigned-off-by: Dong Aisheng \u003cb29396@freescale.com\u003e\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "921f168109d029e3b59459c3a2754f0e2a70fad3", "tree": "9cae8a1f1abfbf8d3e88e4fd1ba2f8df37b02d34", "parents": [ "f6a99649526370c7a88b86736c02f2a74b5b20e7" ], "author": { "name": "Dong Aisheng", "email": "b29396@freescale.com", "time": "Tue Nov 18 19:00:54 2014 +0800" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:35:04 2014 +0100" }, "message": "can: m_can: fix not set can_dlc for remote frame\n\nThe original code missed to set the cf-\u003ecan_dlc in the RTR case, so add it.\n\nSigned-off-by: Dong Aisheng \u003cb29396@freescale.com\u003e\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "f6a99649526370c7a88b86736c02f2a74b5b20e7", "tree": "f048cdaebc0c5ac43264ba7b2ae6079e75086a17", "parents": [ "962845da54afe2fc7ebf64c2d8bb5aa65a826b2f" ], "author": { "name": "Dong Aisheng", "email": "b29396@freescale.com", "time": "Wed Oct 29 18:45:21 2014 +0800" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:35:03 2014 +0100" }, "message": "can: m_can: fix possible sleep in napi poll\n\nThe m_can_get_berr_counter function can sleep and it may be called in napi poll\nfunction. Rework it to fix the following warning.\n\nroot@imx6qdlsolo:~# cangen can0 -f -L 12 -D 112233445566778899001122\n[ 1846.017565] m_can 20e8000.can can0: entered error warning state\n[ 1846.023551] ------------[ cut here ]------------\n[ 1846.028216] WARNING: CPU: 0 PID: 560 at kernel/locking/mutex.c:867 mutex_trylock+0x218/0x23c()\n[ 1846.036889] DEBUG_LOCKS_WARN_ON(in_interrupt())\n[ 1846.041263] Modules linked in:\n[ 1846.044594] CPU: 0 PID: 560 Comm: cangen Not tainted 3.17.0-rc4-next-20140915-00010-g032d018-dirty #477\n[ 1846.054033] Backtrace:\n[ 1846.056557] [\u003c80012448\u003e] (dump_backtrace) from [\u003c80012728\u003e] (show_stack+0x18/0x1c)\n[ 1846.064180] r6:809a07ec r5:809a07ec r4:00000000 r3:00000000\n[ 1846.069966] [\u003c80012710\u003e] (show_stack) from [\u003c806c9ee0\u003e] (dump_stack+0x8c/0xa4)\n[ 1846.077264] [\u003c806c9e54\u003e] (dump_stack) from [\u003c8002aa78\u003e] (warn_slowpath_common+0x70/0x94)\n[ 1846.085403] r6:806cd1b0 r5:00000009 r4:be1d5c20 r3:be07b0c0\n[ 1846.091204] [\u003c8002aa08\u003e] (warn_slowpath_common) from [\u003c8002aad4\u003e] (warn_slowpath_fmt+0x38/0x40)\n[ 1846.099951] r8:8119106c r7:80515aa4 r6:be027000 r5:00000001 r4:809d1df4\n[ 1846.106830] [\u003c8002aaa0\u003e] (warn_slowpath_fmt) from [\u003c806cd1b0\u003e] (mutex_trylock+0x218/0x23c)\n[ 1846.115141] r3:80851c88 r2:8084fb74\n[ 1846.118804] [\u003c806ccf98\u003e] (mutex_trylock) from [\u003c80515aa4\u003e] (clk_prepare_lock+0x14/0xf4)\n[ 1846.126859] r8:00000040 r7:be1d5cec r6:be027000 r5:be255800 r4:be027000\n[ 1846.133737] [\u003c80515a90\u003e] (clk_prepare_lock) from [\u003c80517660\u003e] (clk_prepare+0x14/0x2c)\n[ 1846.141583] r5:be255800 r4:be027000\n[ 1846.145272] [\u003c8051764c\u003e] (clk_prepare) from [\u003c8041ff14\u003e] (m_can_get_berr_counter+0x20/0xd4)\n[ 1846.153672] r4:be255800 r3:be07b0c0\n[ 1846.157325] [\u003c8041fef4\u003e] (m_can_get_berr_counter) from [\u003c80420428\u003e] (m_can_poll+0x310/0x8fc)\n[ 1846.165809] r7:bd4dc540 r6:00000744 r5:11300000 r4:be255800\n[ 1846.171590] [\u003c80420118\u003e] (m_can_poll) from [\u003c8056a468\u003e] (net_rx_action+0xcc/0x1b4)\n[ 1846.179204] r10:00000101 r9:be255ebc r8:00000040 r7:be7c3208 r6:8097c100 r5:be7c3200\n[ 1846.187192] r4:0000012c\n[ 1846.189779] [\u003c8056a39c\u003e] (net_rx_action) from [\u003c8002deec\u003e] (__do_softirq+0xfc/0x2c4)\n[ 1846.197568] r10:00000101 r9:8097c088 r8:00000003 r7:8097c080 r6:40000001 r5:8097c08c\n[ 1846.205559] r4:00000020\n[ 1846.208144] [\u003c8002ddf0\u003e] (__do_softirq) from [\u003c8002e194\u003e] (do_softirq+0x7c/0x88)\n[ 1846.215588] r10:00000000 r9:bd516a60 r8:be18ce00 r7:00000000 r6:be255800 r5:8056c0ec\n[ 1846.223578] r4:60000093\n[ 1846.226163] [\u003c8002e118\u003e] (do_softirq) from [\u003c8002e288\u003e] (__local_bh_enable_ip+0xe8/0x10c)\n[ 1846.234386] r4:00000200 r3:be1d4000\n[ 1846.238036] [\u003c8002e1a0\u003e] (__local_bh_enable_ip) from [\u003c8056c108\u003e] (__dev_queue_xmit+0x314/0x6b0)\n[ 1846.246868] r6:be255800 r5:bd516a00 r4:00000000 r3:be07b0c0\n[ 1846.252645] [\u003c8056bdf4\u003e] (__dev_queue_xmit) from [\u003c8056c4b8\u003e] (dev_queue_xmit+0x14/0x18)\n\nSigned-off-by: Dong Aisheng \u003cb29396@freescale.com\u003e\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "962845da54afe2fc7ebf64c2d8bb5aa65a826b2f", "tree": "ed12bbff1d3a12e9861020b5ec7ff3f52ed2d356", "parents": [ "efe22286e05751e9913a8002fefdb45cdb7361ad" ], "author": { "name": "Dong Aisheng", "email": "b29396@freescale.com", "time": "Fri Nov 07 16:45:14 2014 +0800" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:35:03 2014 +0100" }, "message": "can: m_can: add missing message RAM initialization\n\nThe M_CAN message RAM is usually equipped with a parity or ECC functionality.\nBut RAM cells suffer a hardware reset and can therefore hold arbitrary content\nat startup - including parity and/or ECC bits.\n\nTo prevent the M_CAN controller detecting checksum errors when reading\npotentially uninitialized TX message RAM content to transmit CAN frames the TX\nmessage RAM has to be written with (any kind of) initial data.\n\nSigned-off-by: Dong Aisheng \u003cb29396@freescale.com\u003e\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "efe22286e05751e9913a8002fefdb45cdb7361ad", "tree": "5e98ac08639383777ff42dc840608c81b570afb0", "parents": [ "d6fdb38b4f2c4090e176aa55fe73dd2f45cfa4a6" ], "author": { "name": "David Cohen", "email": "david.a.cohen@linux.intel.com", "time": "Wed Oct 15 14:41:50 2014 -0700" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:35:02 2014 +0100" }, "message": "can: m_can: add CONFIG_HAS_IOMEM dependence\n\nm_can uses io memory which makes it not compilable on architectures\nwithout HAS_IOMEM such as UML:\n\ndrivers/built-in.o: In function `m_can_plat_probe\u0027:\nm_can.c:(.text+0x218cc5): undefined reference to `devm_ioremap_resource\u0027\nm_can.c:(.text+0x218df9): undefined reference to `devm_ioremap\u0027\n\nSigned-off-by: David Cohen \u003cdavid.a.cohen@linux.intel.com\u003e\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "d6fdb38b4f2c4090e176aa55fe73dd2f45cfa4a6", "tree": "27135041d0f5c1121abbd90d44eb761adfe2328e", "parents": [ "50212b425d0ef8b606b316826b56abbb48680758" ], "author": { "name": "Dong Aisheng", "email": "b29396@freescale.com", "time": "Wed Oct 29 18:45:23 2014 +0800" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:35:02 2014 +0100" }, "message": "can: m_can: add .ndo_change_mtu function\n\nUse common can_change_mtu function.\n\nSigned-off-by: Dong Aisheng \u003cb29396@freescale.com\u003e\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "50212b425d0ef8b606b316826b56abbb48680758", "tree": "3ada0d7526ec0a5d87570ca83fd6b861cba8d452", "parents": [ "ca976d6af4e50899f4e840e8017ec29d8fe6b50e" ], "author": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 13:16:13 2014 +0100" }, "committer": { "name": "Marc Kleine-Budde", "email": "mkl@pengutronix.de", "time": "Tue Nov 18 21:34:59 2014 +0100" }, "message": "can: gs_usb: add .ndo_change_mtu function\n\nUse common can_change_mtu function.\n\nSigned-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n" }, { "commit": "a0e27f51ba8a04125c22a95c4d3e98297a7191de", "tree": "b021e92474ccff983ffe32e3c3618c3701646b1f", "parents": [ "5641c09226f401ee054e48521707fb185380e8d3" ], "author": { "name": "Soren Brinkmann", "email": "soren.brinkmann@xilinx.com", "time": "Thu Nov 06 07:38:51 2014 -0800" }, "committer": { "name": "Rob Herring", "email": "robh@kernel.org", "time": "Tue Nov 18 12:01:52 2014 -0600" }, "message": "documentation: pinctrl bindings: Fix trivial typo \u0027abitrary\u0027\n\nA misspelled \u0027arbitrary\u0027 propagated to quite a few locations in the DT\nbinding documentation for pin-controllers. Fixing by:\n git grep abitrary | cut -f1 -d: | xargs sed -i \u0027s/abitrary/arbitrary/\u0027\n\nReported-by: Andreas Färber \u003cafaerber@suse.de\u003e\nSigned-off-by: Soren Brinkmann \u003csoren.brinkmann@xilinx.com\u003e\nSigned-off-by: Rob Herring \u003crobh@kernel.org\u003e\n" }, { "commit": "5641c09226f401ee054e48521707fb185380e8d3", "tree": "774bc1aa6f83ca3ea3780b8e06117fabbd758928", "parents": [ "f9cb89b63db8cb2755a5179843a0643cc284f1ef" ], "author": { "name": "bpqw", "email": "bpqw@micron.com", "time": "Wed Nov 12 14:26:42 2014 +0000" }, "committer": { "name": "Rob Herring", "email": "robh@kernel.org", "time": "Tue Nov 18 11:51:35 2014 -0600" }, "message": "devicetree: bindings: Add vendor prefix for Micron Technology, Inc.\n\nThis patch is used to add vendor prefix for Micron Technology, Inc. in\nthe vendor-prefixes.txt file.\n\nMicron Technology, Inc. is an American multinational corporation based\nin Boise, Idaho, best known for producing many forms of semiconductor\ndevices. This includes DRAM, SDRAM, flash memory, eMMC and SSDs.\n\nSigned-off-by: Bean Huo \u003cbpqw@micron.com\u003e\n[robh: cleanup commit msg formatting and company name]\nSigned-off-by: Rob Herring \u003crobh@kernel.org\u003e\n" }, { "commit": "f9cb89b63db8cb2755a5179843a0643cc284f1ef", "tree": "846b3fa379d86482efc2abb2bd2e97905f3f4594", "parents": [ "746c9e9f92dde2789908e51a354ba90a1962a2eb" ], "author": { "name": "Philipp Zabel", "email": "p.zabel@pengutronix.de", "time": "Wed May 14 11:24:43 2014 +0200" }, "committer": { "name": "Rob Herring", "email": "robh@kernel.org", "time": "Tue Nov 18 11:46:44 2014 -0600" }, "message": "of: Add vendor prefix for Chips\u0026Media, Inc.\n\nChips\u0026Media is a developer of Video Codec IP cores.\n\nSigned-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n[robh: fix-up alphabetical ordering]\nSigned-off-by: Rob Herring \u003crobh@kernel.org\u003e\n" }, { "commit": "746c9e9f92dde2789908e51a354ba90a1962a2eb", "tree": "5ca9d6602cd4eb71e1d5828b2a52666060a32f05", "parents": [ "9b6eab07588c2de102423fe99c875fc4bfda2508" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Fri Nov 14 17:55:03 2014 +1100" }, "committer": { "name": "Rob Herring", "email": "robh@kernel.org", "time": "Tue Nov 18 11:46:44 2014 -0600" }, "message": "of/base: Fix PowerPC address parsing hack\n\nWe have a historical hack that treats missing ranges properties as the\nequivalent of an empty one. This is needed for ancient PowerMac \"bad\"\ndevice-trees, and shouldn\u0027t be enabled for any other PowerPC platform,\notherwise we get some nasty layout of devices in sysfs or even\nduplication when a set of otherwise identically named devices is\ncreated multiple times under a different parent node with no ranges\nproperty.\n\nThis fix is needed for the PowerNV i2c busses to be exposed properly\nand will fix a number of other embedded cases.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCC: \u003cstable@vger.kernel.org\u003e\nAcked-by: Grant Likely \u003cgrant.likely@linaro.org\u003e\nSigned-off-by: Rob Herring \u003crobh@kernel.org\u003e\n" } ], "next": "9b6eab07588c2de102423fe99c875fc4bfda2508" }