)]}' { "log": [ { "commit": "d20810530b7109a95abef5130e6dcec09c5180d7", "tree": "6617c73b795f51f1e86a6cdb79fadb7ed5d69829", "parents": [ "64999fa7aa2c076ec6d05aee481f11f5296ceb8c" ], "author": { "name": "Brajeswar Ghosh", "email": "brajeswar.linux@gmail.com", "time": "Fri Nov 16 16:17:03 2018 +0530" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Nov 26 10:13:04 2018 +1100" }, "message": "fsi: fsi-scom.c: Remove duplicate header\n\nRemove linux/cdev.h which is included more than once\n\nSigned-off-by: Brajeswar Ghosh \u003cbrajeswar.linux@gmail.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "64999fa7aa2c076ec6d05aee481f11f5296ceb8c", "tree": "b0e1bd8c512c6a5b4f950aba78b3d74d2ef4d2e9", "parents": [ "2e6e902d185027f8e3cb8b7305238f7e35d6a436" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Tue Aug 14 00:37:18 2018 +0200" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Nov 26 10:11:43 2018 +1100" }, "message": "fsi: master-ast-cf: select GENERIC_ALLOCATOR\n\nIn randconfig builds without CONFIG_GENERIC_ALLOCATOR, this driver\nfails to link:\n\nERROR: \"gen_pool_alloc_algo\" [drivers/fsi/fsi-master-ast-cf.ko] undefined!\nERROR: \"gen_pool_fixed_alloc\" [drivers/fsi/fsi-master-ast-cf.ko] undefined!\nERROR: \"of_gen_pool_get\" [drivers/fsi/fsi-master-ast-cf.ko] undefined!\nERROR: \"gen_pool_free\" [drivers/fsi/fsi-master-ast-cf.ko] undefined!\n\nSelect the dependency as all other users do.\n\nFixes: 6a794a27daca (\"fsi: master-ast-cf: Add new FSI master using Aspeed ColdFire\")\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "2e6e902d185027f8e3cb8b7305238f7e35d6a436", "tree": "9b3ba6d8ff82308734c5d57f7d3957f62466b304", "parents": [ "d6d460b89378b1bc6715574cdafd748ba59d5a27" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 25 14:19:31 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 25 14:19:31 2018 -0800" }, "message": "Linux 4.20-rc4\n" }, { "commit": "d6d460b89378b1bc6715574cdafd748ba59d5a27", "tree": "b43a2f09bf1fa86264cc01c42f0650832c6619f0", "parents": [ "17c2f540863a6c0faa3f0ede3c785d9427bcaf80", "cb216b84d6ea24fa10f1e7aac35de77246841041" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 25 09:24:40 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 25 09:24:40 2018 -0800" }, "message": "Merge tag \u0027dma-mapping-4.20-3\u0027 of git://git.infradead.org/users/hch/dma-mapping\n\nPull dma-mapping fixes from Christoph Hellwig:\n \"Two dma-direct / swiotlb regressions fixes:\n\n - zero is a valid physical address on some arm boards, we can\u0027t use\n it as the error value\n\n - don\u0027t try to cache flush the error return value (no matter what it\n is)\"\n\n* tag \u0027dma-mapping-4.20-3\u0027 of git://git.infradead.org/users/hch/dma-mapping:\n swiotlb: Skip cache maintenance on map error\n dma-direct: Make DIRECT_MAPPING_ERROR viable for SWIOTLB\n" }, { "commit": "17c2f540863a6c0faa3f0ede3c785d9427bcaf80", "tree": "8c90950536b5bbbd63783b1b29b90f7e1907c647", "parents": [ "4e962ff6e34f44c400c548da0c1e2393053a691e", "bb21ce0ad227b69ec0f83279297ee44232105d96" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 25 09:19:58 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 25 09:19:58 2018 -0800" }, "message": "Merge tag \u0027nfs-for-4.20-4\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs\n\nPull NFS client bugfixes from Trond Myklebust:\n\n - Fix a NFSv4 state manager deadlock when returning a delegation\n\n - NFSv4.2 copy do not allocate memory under the lock\n\n - flexfiles: Use the correct stateid for IO in the tightly coupled case\n\n* tag \u0027nfs-for-4.20-4\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs:\n flexfiles: use per-mirror specified stateid for IO\n NFSv4.2 copy do not allocate memory under the lock\n NFSv4: Fix a NFSv4 state manager deadlock\n" }, { "commit": "4e962ff6e34f44c400c548da0c1e2393053a691e", "tree": "dc6750adfd23a9a12af5f74fcb59a43363afaee9", "parents": [ "e2125dac22f2c9c66c412cd8e049a7305af59f73" ], "author": { "name": "Luc Van Oostenryck", "email": "luc.vanoostenryck@gmail.com", "time": "Sun Nov 25 17:34:05 2018 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Nov 25 09:17:43 2018 -0800" }, "message": "MAINTAINERS: change Sparse\u0027s maintainer\n\nI\u0027m taking over the maintainance of Sparse so add myself as\nmaintainer and move Christopher\u0027s info to CREDITS.\n\nSigned-off-by: Luc Van Oostenryck \u003cluc.vanoostenryck@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e2125dac22f2c9c66c412cd8e049a7305af59f73", "tree": "8584af248ee592622bd6e8a7f9cee0b041b784cc", "parents": [ "e195ca6cb6f21633e56322d5aa11ed59cdb22fb2", "fffc9a260e38acec3187515738122a3ecb24ac90" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 18:44:01 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 18:44:01 2018 -0800" }, "message": "Merge tag \u0027xarray-4.20-rc4\u0027 of git://git.infradead.org/users/willy/linux-dax\n\nPull XArray updates from Matthew Wilcox:\n \"We found some bugs in the DAX conversion to XArray (and one bug which\n predated the XArray conversion). There were a couple of bugs in some\n of the higher-level functions, which aren\u0027t actually being called in\n today\u0027s kernel, but surfaced as a result of converting existing radix\n tree \u0026 IDR users over to the XArray.\n\n Some of the other changes to how the higher-level APIs work were also\n motivated by converting various users; again, they\u0027re not in use in\n today\u0027s kernel, so changing them has a low probability of introducing\n a bug.\n\n Dan can still trigger a bug in the DAX code with hot-offline/online,\n and we\u0027re working on tracking that down\"\n\n* tag \u0027xarray-4.20-rc4\u0027 of git://git.infradead.org/users/willy/linux-dax:\n XArray tests: Add missing locking\n dax: Avoid losing wakeup in dax_lock_mapping_entry\n dax: Fix huge page faults\n dax: Fix dax_unlock_mapping_entry for PMD pages\n dax: Reinstate RCU protection of inode\n dax: Make sure the unlocking entry isn\u0027t locked\n dax: Remove optimisation from dax_lock_mapping_entry\n XArray tests: Correct some 64-bit assumptions\n XArray: Correct xa_store_range\n XArray: Fix Documentation\n XArray: Handle NULL pointers differently for allocation\n XArray: Unify xa_store and __xa_store\n XArray: Add xa_store_bh() and xa_store_irq()\n XArray: Turn xa_erase into an exported function\n XArray: Unify xa_cmpxchg and __xa_cmpxchg\n XArray: Regularise xa_reserve\n nilfs2: Use xa_erase_irq\n XArray: Export __xa_foo to non-GPL modules\n XArray: Fix xa_for_each with a single element at 0\n" }, { "commit": "e195ca6cb6f21633e56322d5aa11ed59cdb22fb2", "tree": "63f4a3ccc0e16f9c4b3d86a1e487da41394af20d", "parents": [ "d146194f31c96f9b260c5a1cf1592d2e7f82a2e2", "ffe0e7cf290f5c9d1392134b4ef8da2a3761a4cd" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 12:58:47 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 12:58:47 2018 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid\n\nPull HID fixes from Jiri Kosina:\n\n - revert of the high-resolution scrolling feature, as it breaks certain\n hardware due to incompatibilities between Logitech and Microsoft\n worlds. Peter Hutterer is working on a fixed implementation. Until\n that is finished, revert by Benjamin Tissoires.\n\n - revert of incorrect strncpy-\u003estrlcpy conversion in uhid, from David\n Herrmann\n\n - fix for buggy sendfile() implementation on uhid device node, from\n Eric Biggers\n\n - a few assorted device-ID specific quirks\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid:\n Revert \"Input: Add the `REL_WHEEL_HI_RES` event code\"\n Revert \"HID: input: Create a utility class for counting scroll events\"\n Revert \"HID: logitech: Add function to enable HID++ 1.0 \"scrolling acceleration\"\"\n Revert \"HID: logitech: Enable high-resolution scrolling on Logitech mice\"\n Revert \"HID: logitech: Use LDJ_DEVICE macro for existing Logitech mice\"\n Revert \"HID: logitech: fix a used uninitialized GCC warning\"\n Revert \"HID: input: simplify/fix high-res scroll event handling\"\n HID: Add quirk for Primax PIXART OEM mice\n HID: i2c-hid: Disable runtime PM for LG touchscreen\n HID: multitouch: Add pointstick support for Cirque Touchpad\n HID: steam: remove input device when a hid client is running.\n Revert \"HID: uhid: use strlcpy() instead of strncpy()\"\n HID: uhid: forbid UHID_CREATE under KERNEL_DS or elevated privileges\n HID: input: Ignore battery reported by Symbol DS4308\n HID: Add quirk for Microsoft PIXART OEM mouse\n" }, { "commit": "d146194f31c96f9b260c5a1cf1592d2e7f82a2e2", "tree": "ff576fb60ffbe6d3bcc325b217aef65e05c1331a", "parents": [ "857fa628bbe93017c72ddd0d5304962a2608db07", "4f9f49646a5733c0c2bd49940673dde89a9c5add" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 09:42:32 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 09:42:32 2018 -0800" }, "message": "Merge tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux\n\nPull arm64 fixes from Catalin Marinas::\n\n - Fix wrong conflict resolution around CONFIG_ARM64_SSBD\n\n - Fix sparse warning on unsigned long constant\n\n* tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:\n arm64: cpufeature: Fix mismerge of CONFIG_ARM64_SSBD block\n arm64: sysreg: fix sparse warnings\n" }, { "commit": "857fa628bbe93017c72ddd0d5304962a2608db07", "tree": "30e0afe2865fb15b357559f342c18139347f24f0", "parents": [ "abe72ff4134028ff2189d29629c40a40bee0a989", "07093b76476903f820d83d56c3040e656fb4d9e3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 09:19:38 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 09:19:38 2018 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net\n\nPull networking fixes from David Miller:\n\n 1) Need to take mutex in ath9k_add_interface(), from Dan Carpenter.\n\n 2) Fix mt76 build without CONFIG_LEDS_CLASS, from Arnd Bergmann.\n\n 3) Fix socket wmem accounting in SCTP, from Xin Long.\n\n 4) Fix failed resume crash in ena driver, from Arthur Kiyanovski.\n\n 5) qed driver passes bytes instead of bits into second arg of\n bitmap_weight(). From Denis Bolotin.\n\n 6) Fix reset deadlock in ibmvnic, from Juliet Kim.\n\n 7) skb_scrube_packet() needs to scrub the fwd marks too, from Petr\n Machata.\n\n 8) Make sure older TCP stacks see enough dup ACKs, and avoid doing SACK\n compression during this period, from Eric Dumazet.\n\n 9) Add atomicity to SMC protocol cursor handling, from Ursula Braun.\n\n10) Don\u0027t leave dangling error pointer if bpf_prog_add() fails in\n thunderx driver, from Lorenzo Bianconi. Also, when we unmap TSO\n headers, set sq-\u003etso_hdrs to NULL.\n\n11) Fix race condition over state variables in act_police, from Davide\n Caratti.\n\n12) Disable guest csum in the presence of XDP in virtio_net, from Jason\n Wang.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (64 commits)\n net: gemini: Fix copy/paste error\n net: phy: mscc: fix deadlock in vsc85xx_default_config\n dt-bindings: dsa: Fix typo in \"probed\"\n net: thunderx: set tso_hdrs pointer to NULL in nicvf_free_snd_queue\n net: amd: add missing of_node_put()\n team: no need to do team_notify_peers or team_mcast_rejoin when disabling port\n virtio-net: fail XDP set if guest csum is negotiated\n virtio-net: disable guest csum during XDP set\n net/sched: act_police: add missing spinlock initialization\n net: don\u0027t keep lonely packets forever in the gro hash\n net/ipv6: re-do dad when interface has IFF_NOARP flag change\n packet: copy user buffers before orphan or clone\n ibmvnic: Update driver queues after change in ring size support\n ibmvnic: Fix RX queue buffer cleanup\n net: thunderx: set xdp_prog to NULL if bpf_prog_add fails\n net/dim: Update DIM start sample after each DIM iteration\n net: faraday: ftmac100: remove netif_running(netdev) check before disabling interrupts\n net/smc: use after free fix in smc_wr_tx_put_slot()\n net/smc: atomic SMCD cursor handling\n net/smc: add SMC-D shutdown signal\n ...\n" }, { "commit": "abe72ff4134028ff2189d29629c40a40bee0a989", "tree": "131d5cb0b4f091ac963c8a444098bec2025577cd", "parents": [ "7c98a42618271210c60b79128b220107d35938d9", "8c110d43c6bca4b24dd13272a9d4e0ba6f2ec957" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 09:11:52 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Nov 24 09:11:52 2018 -0800" }, "message": "Merge tag \u0027xfs-4.20-fixes-2\u0027 of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux\n\nPull xfs fixes from Darrick Wong:\n \"Dave and I have continued our work fixing corruption problems that can\n be found when running long-term burn-in exercisers on xfs. Here are\n some patches fixing most of the problems, but there will likely be\n more. :/\n\n - Numerous corruption fixes for copy on write\n\n - Numerous corruption fixes for blocksize \u003c pagesize writes\n\n - Don\u0027t miscalculate AG reservations for small final AGs\n\n - Fix page cache truncation to work properly for reflink and extent\n shifting\n\n - Fix use-after-free when retrying failed inode/dquot buffer logging\n\n - Fix corruptions seen when using copy_file_range in directio mode\"\n\n* tag \u0027xfs-4.20-fixes-2\u0027 of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:\n iomap: readpages doesn\u0027t zero page tail beyond EOF\n vfs: vfs_dedupe_file_range() doesn\u0027t return EOPNOTSUPP\n iomap: dio data corruption and spurious errors when pipes fill\n iomap: sub-block dio needs to zeroout beyond EOF\n iomap: FUA is wrong for DIO O_DSYNC writes into unwritten extents\n xfs: delalloc -\u003e unwritten COW fork allocation can go wrong\n xfs: flush removing page cache in xfs_reflink_remap_prep\n xfs: extent shifting doesn\u0027t fully invalidate page cache\n xfs: finobt AG reserves don\u0027t consider last AG can be a runt\n xfs: fix transient reference count error in xfs_buf_resubmit_failed_buffers\n xfs: uncached buffer tracing needs to print bno\n xfs: make xfs_file_remap_range() static\n xfs: fix shared extent data corruption due to missing cow reservation\n" }, { "commit": "07093b76476903f820d83d56c3040e656fb4d9e3", "tree": "cd494c972b26cdc54ad16139935f80e9ee5c84a7", "parents": [ "3fa528b7682e73e906266bcd43728b8f923bf9b2" ], "author": { "name": "Andreas Fiedler", "email": "andreas.fiedler@gmx.net", "time": "Sat Nov 24 00:16:34 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 22:35:38 2018 -0800" }, "message": "net: gemini: Fix copy/paste error\n\nThe TX stats should be started with the tx_stats_syncp,\nthere seems to be a copy/paste error in the driver.\n\nSigned-off-by: Andreas Fiedler \u003candreas.fiedler@gmx.net\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3fa528b7682e73e906266bcd43728b8f923bf9b2", "tree": "57494784e7f165aa9f445ecef2d202cb2f04e662", "parents": [ "e7b9fb4f545b1f7885e7c642643828f93d3d79c9" ], "author": { "name": "Quentin Schulz", "email": "quentin.schulz@bootlin.com", "time": "Fri Nov 23 19:01:51 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 22:34:40 2018 -0800" }, "message": "net: phy: mscc: fix deadlock in vsc85xx_default_config\n\nThe vsc85xx_default_config function called in the vsc85xx_config_init\nfunction which is used by VSC8530, VSC8531, VSC8540 and VSC8541 PHYs\nmistakenly calls phy_read and phy_write in-between phy_select_page and\nphy_restore_page.\n\nphy_select_page and phy_restore_page actually take and release the MDIO\nbus lock and phy_write and phy_read take and release the lock to write\nor read to a PHY register.\n\nLet\u0027s fix this deadlock by using phy_modify_paged which handles\ncorrectly a read followed by a write in a non-standard page.\n\nFixes: 6a0bfbbe20b0 (\"net: phy: mscc: migrate to phy_select/restore_page functions\")\nSigned-off-by: Quentin Schulz \u003cquentin.schulz@bootlin.com\u003e\nReviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e7b9fb4f545b1f7885e7c642643828f93d3d79c9", "tree": "79dc4755129d694242d76d4f1c5fda353ccad064", "parents": [ "ef2a7cf1d8831535b8991459567b385661eb4a36" ], "author": { "name": "Fabio Estevam", "email": "festevam@gmail.com", "time": "Fri Nov 23 15:46:50 2018 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 22:33:55 2018 -0800" }, "message": "dt-bindings: dsa: Fix typo in \"probed\"\n\nThe correct form is \"can be probed\", so fix the typo.\n\nSigned-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\nReviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ef2a7cf1d8831535b8991459567b385661eb4a36", "tree": "aec4ed1689b5ab2860c91d956813706a3a62842c", "parents": [ "c44c749d3b6fdfca39002e7e48e03fe9f9fe37a3" ], "author": { "name": "Lorenzo Bianconi", "email": "lorenzo.bianconi@redhat.com", "time": "Fri Nov 23 18:28:01 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 22:31:56 2018 -0800" }, "message": "net: thunderx: set tso_hdrs pointer to NULL in nicvf_free_snd_queue\n\nReset snd_queue tso_hdrs pointer to NULL in nicvf_free_snd_queue routine\nsince it is used to check if tso dma descriptor queue has been previously\nallocated. The issue can be triggered with the following reproducer:\n\n$ip link set dev enP2p1s0v0 xdpdrv obj xdp_dummy.o\n$ip link set dev enP2p1s0v0 xdpdrv off\n\n[ 341.467649] WARNING: CPU: 74 PID: 2158 at mm/vmalloc.c:1511 __vunmap+0x98/0xe0\n[ 341.515010] Hardware name: GIGABYTE H270-T70/MT70-HD0, BIOS T49 02/02/2018\n[ 341.521874] pstate: 60400005 (nZCv daif +PAN -UAO)\n[ 341.526654] pc : __vunmap+0x98/0xe0\n[ 341.530132] lr : __vunmap+0x98/0xe0\n[ 341.533609] sp : ffff00001c5db860\n[ 341.536913] x29: ffff00001c5db860 x28: 0000000000020000\n[ 341.542214] x27: ffff810feb5090b0 x26: ffff000017e57000\n[ 341.547515] x25: 0000000000000000 x24: 00000000fbd00000\n[ 341.552816] x23: 0000000000000000 x22: ffff810feb5090b0\n[ 341.558117] x21: 0000000000000000 x20: 0000000000000000\n[ 341.563418] x19: ffff000017e57000 x18: 0000000000000000\n[ 341.568719] x17: 0000000000000000 x16: 0000000000000000\n[ 341.574020] x15: 0000000000000010 x14: ffffffffffffffff\n[ 341.579321] x13: ffff00008985eb27 x12: ffff00000985eb2f\n[ 341.584622] x11: ffff0000096b3000 x10: ffff00001c5db510\n[ 341.589923] x9 : 00000000ffffffd0 x8 : ffff0000086868e8\n[ 341.595224] x7 : 3430303030303030 x6 : 00000000000006ef\n[ 341.600525] x5 : 00000000003fffff x4 : 0000000000000000\n[ 341.605825] x3 : 0000000000000000 x2 : ffffffffffffffff\n[ 341.611126] x1 : ffff0000096b3728 x0 : 0000000000000038\n[ 341.616428] Call trace:\n[ 341.618866] __vunmap+0x98/0xe0\n[ 341.621997] vunmap+0x3c/0x50\n[ 341.624961] arch_dma_free+0x68/0xa0\n[ 341.628534] dma_direct_free+0x50/0x80\n[ 341.632285] nicvf_free_resources+0x160/0x2d8 [nicvf]\n[ 341.637327] nicvf_config_data_transfer+0x174/0x5e8 [nicvf]\n[ 341.642890] nicvf_stop+0x298/0x340 [nicvf]\n[ 341.647066] __dev_close_many+0x9c/0x108\n[ 341.650977] dev_close_many+0xa4/0x158\n[ 341.654720] rollback_registered_many+0x140/0x530\n[ 341.659414] rollback_registered+0x54/0x80\n[ 341.663499] unregister_netdevice_queue+0x9c/0xe8\n[ 341.668192] unregister_netdev+0x28/0x38\n[ 341.672106] nicvf_remove+0xa4/0xa8 [nicvf]\n[ 341.676280] nicvf_shutdown+0x20/0x30 [nicvf]\n[ 341.680630] pci_device_shutdown+0x44/0x88\n[ 341.684720] device_shutdown+0x144/0x250\n[ 341.688640] kernel_restart_prepare+0x44/0x50\n[ 341.692986] kernel_restart+0x20/0x68\n[ 341.696638] __se_sys_reboot+0x210/0x238\n[ 341.700550] __arm64_sys_reboot+0x24/0x30\n[ 341.704555] el0_svc_handler+0x94/0x110\n[ 341.708382] el0_svc+0x8/0xc\n[ 341.711252] ---[ end trace 3f4019c8439959c9 ]---\n[ 341.715874] page:ffff7e0003ef4000 count:0 mapcount:0 mapping:0000000000000000 index:0x4\n[ 341.723872] flags: 0x1fffe000000000()\n[ 341.727527] raw: 001fffe000000000 ffff7e0003f1a008 ffff7e0003ef4048 0000000000000000\n[ 341.735263] raw: 0000000000000004 0000000000000000 00000000ffffffff 0000000000000000\n[ 341.742994] page dumped because: VM_BUG_ON_PAGE(page_ref_count(page) \u003d\u003d 0)\n\nwhere xdp_dummy.c is a simple bpf program that forwards the incoming\nframes to the network stack (available here:\nhttps://github.com/altoor/xdp_walkthrough_examples/blob/master/sample_1/xdp_dummy.c)\n\nFixes: 05c773f52b96 (\"net: thunderx: Add basic XDP support\")\nFixes: 4863dea3fab0 (\"net: Adding support for Cavium ThunderX network controller\")\nSigned-off-by: Lorenzo Bianconi \u003clorenzo.bianconi@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c44c749d3b6fdfca39002e7e48e03fe9f9fe37a3", "tree": "8b1eeb5a4a237d74747834d76df2ad80ceb74a89", "parents": [ "5ed9dc99107144f83b6c1bb52a69b58875baf540" ], "author": { "name": "Yangtao Li", "email": "tiny.windzz@gmail.com", "time": "Thu Nov 22 07:34:41 2018 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 17:24:24 2018 -0800" }, "message": "net: amd: add missing of_node_put()\n\nof_find_node_by_path() acquires a reference to the node\nreturned by it and that reference needs to be dropped by its caller.\nThis place doesn\u0027t do that, so fix it.\n\nSigned-off-by: Yangtao Li \u003ctiny.windzz@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5ed9dc99107144f83b6c1bb52a69b58875baf540", "tree": "438ec2a7d4d3fc051ed9ecf1581439f59c081c70", "parents": [ "18ba58e1c234ea1a2d9835ac8c1735d965ce4640" ], "author": { "name": "Hangbin Liu", "email": "liuhangbin@gmail.com", "time": "Thu Nov 22 16:15:28 2018 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 17:18:15 2018 -0800" }, "message": "team: no need to do team_notify_peers or team_mcast_rejoin when disabling port\n\nteam_notify_peers() will send ARP and NA to notify peers. team_mcast_rejoin()\nwill send multicast join group message to notify peers. We should do this when\nenabling/changed to a new port. But it doesn\u0027t make sense to do it when a port\nis disabled.\n\nOn the other hand, when we set mcast_rejoin_count to 2, and do a failover,\nteam_port_disable() will increase mcast_rejoin.count_pending to 2 and then\nteam_port_enable() will increase mcast_rejoin.count_pending to 4. We will send\n4 mcast rejoin messages at latest, which will make user confused. The same\nwith notify_peers.count.\n\nFix it by deleting team_notify_peers() and team_mcast_rejoin() in\nteam_port_disable().\n\nReported-by: Liang Li \u003cliali@redhat.com\u003e\nFixes: fc423ff00df3a (\"team: add peer notification\")\nFixes: 492b200efdd20 (\"team: add support for sending multicast rejoins\")\nSigned-off-by: Hangbin Liu \u003cliuhangbin@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "18ba58e1c234ea1a2d9835ac8c1735d965ce4640", "tree": "a6d6ca875bf06a4d3213108dbaddbcdee5c18f63", "parents": [ "e59ff2c49ae16e1d179de679aca81405829aee6c" ], "author": { "name": "Jason Wang", "email": "jasowang@redhat.com", "time": "Thu Nov 22 14:36:31 2018 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 11:59:40 2018 -0800" }, "message": "virtio-net: fail XDP set if guest csum is negotiated\n\nWe don\u0027t support partial csumed packet since its metadata will be lost\nor incorrect during XDP processing. So fail the XDP set if guest_csum\nfeature is negotiated.\n\nFixes: f600b6905015 (\"virtio_net: Add XDP support\")\nReported-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nCc: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nCc: Pavel Popa \u003cpashinho1990@gmail.com\u003e\nCc: David Ahern \u003cdsahern@gmail.com\u003e\nSigned-off-by: Jason Wang \u003cjasowang@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e59ff2c49ae16e1d179de679aca81405829aee6c", "tree": "d0da9016181e1806fc2efcca0c0a3b89c2fa9c1e", "parents": [ "484afd1bd3fc6f9f5347289fc8b285aa65f67054" ], "author": { "name": "Jason Wang", "email": "jasowang@redhat.com", "time": "Thu Nov 22 14:36:30 2018 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 11:59:40 2018 -0800" }, "message": "virtio-net: disable guest csum during XDP set\n\nWe don\u0027t disable VIRTIO_NET_F_GUEST_CSUM if XDP was set. This means we\ncan receive partial csumed packets with metadata kept in the\nvnet_hdr. This may have several side effects:\n\n- It could be overridden by header adjustment, thus is might be not\n correct after XDP processing.\n- There\u0027s no way to pass such metadata information through\n XDP_REDIRECT to another driver.\n- XDP does not support checksum offload right now.\n\nSo simply disable guest csum if possible in this the case of XDP.\n\nFixes: 3f93522ffab2d (\"virtio-net: switch off offloads on demand if possible on XDP set\")\nReported-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nCc: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nCc: Pavel Popa \u003cpashinho1990@gmail.com\u003e\nCc: David Ahern \u003cdsahern@gmail.com\u003e\nSigned-off-by: Jason Wang \u003cjasowang@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7c98a42618271210c60b79128b220107d35938d9", "tree": "79ae0c1ca1b98b94f700d28cd6cb50fc80d866ff", "parents": [ "3381918fec9278d14f776d1dabd68da85fd6822e", "7e241f647dc7087a0401418a187f3f5b527cc690" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 11:24:55 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 11:24:55 2018 -0800" }, "message": "Merge tag \u0027ceph-for-4.20-rc4\u0027 of https://github.com/ceph/ceph-client\n\nPullk ceph fix from Ilya Dryomov:\n \"A messenger fix, marked for stable\"\n\n* tag \u0027ceph-for-4.20-rc4\u0027 of https://github.com/ceph/ceph-client:\n libceph: fall back to sendmsg for slab pages\n" }, { "commit": "3381918fec9278d14f776d1dabd68da85fd6822e", "tree": "9ef5806b5ab9f7b941d5069fa9c7acd6d4f901c3", "parents": [ "d88783b9c8849d88c3a75b7b9071cba072b47eba", "14b04063cc994effc86f976625bf8f806d8d44cb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 11:20:14 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 11:20:14 2018 -0800" }, "message": "Merge tag \u0027for-linus-20181123\u0027 of git://git.kernel.dk/linux-block\n\nPull block fix from Jens Axboe:\n \"Just a single fix for this week, fixing an issue with nvme-fc\"\n\n* tag \u0027for-linus-20181123\u0027 of git://git.kernel.dk/linux-block:\n nvme-fc: resolve io failures during connect\n" }, { "commit": "484afd1bd3fc6f9f5347289fc8b285aa65f67054", "tree": "7ff88b693e002f56295a4b0d1878c749cf509fbf", "parents": [ "605108acfe6233b72e2f803aa1cb59a2af3001ca" ], "author": { "name": "Davide Caratti", "email": "dcaratti@redhat.com", "time": "Wed Nov 21 18:23:53 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 11:20:02 2018 -0800" }, "message": "net/sched: act_police: add missing spinlock initialization\n\ncommit f2cbd4852820 (\"net/sched: act_police: fix race condition on state\nvariables\") introduces a new spinlock, but forgets its initialization.\nEnsure that tcf_police_init() initializes \u0027tcfp_lock\u0027 every time a \u0027police\u0027\naction is newly created, to avoid the following lockdep splat:\n\n INFO: trying to register non-static key.\n the code is fine but needs lockdep annotation.\n turning off the locking correctness validator.\n \u003c...\u003e\n Call Trace:\n dump_stack+0x85/0xcb\n register_lock_class+0x581/0x590\n __lock_acquire+0xd4/0x1330\n ? tcf_police_init+0x2fa/0x650 [act_police]\n ? lock_acquire+0x9e/0x1a0\n lock_acquire+0x9e/0x1a0\n ? tcf_police_init+0x2fa/0x650 [act_police]\n ? tcf_police_init+0x55a/0x650 [act_police]\n _raw_spin_lock_bh+0x34/0x40\n ? tcf_police_init+0x2fa/0x650 [act_police]\n tcf_police_init+0x2fa/0x650 [act_police]\n tcf_action_init_1+0x384/0x4c0\n tcf_action_init+0xf6/0x160\n tcf_action_add+0x73/0x170\n tc_ctl_action+0x122/0x160\n rtnetlink_rcv_msg+0x2a4/0x490\n ? netlink_deliver_tap+0x99/0x400\n ? validate_linkmsg+0x370/0x370\n netlink_rcv_skb+0x4d/0x130\n netlink_unicast+0x196/0x230\n netlink_sendmsg+0x2e5/0x3e0\n sock_sendmsg+0x36/0x40\n ___sys_sendmsg+0x280/0x2f0\n ? _raw_spin_unlock+0x24/0x30\n ? handle_pte_fault+0xafe/0xf30\n ? find_held_lock+0x2d/0x90\n ? syscall_trace_enter+0x1df/0x360\n ? __sys_sendmsg+0x5e/0xa0\n __sys_sendmsg+0x5e/0xa0\n do_syscall_64+0x60/0x210\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n RIP: 0033:0x7f1841c7cf10\n Code: c3 48 8b 05 82 6f 2c 00 f7 db 64 89 18 48 83 cb ff eb dd 0f 1f 80 00 00 00 00 83 3d 8d d0 2c 00 00 75 10 b8 2e 00 00 00 0f 05 \u003c48\u003e 3d 01 f0 ff ff 73 31 c3 48 83 ec 08 e8 ae cc 00 00 48 89 04 24\n RSP: 002b:00007ffcf9df4d68 EFLAGS: 00000246 ORIG_RAX: 000000000000002e\n RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007f1841c7cf10\n RDX: 0000000000000000 RSI: 00007ffcf9df4dc0 RDI: 0000000000000003\n RBP: 000000005bf56105 R08: 0000000000000002 R09: 00007ffcf9df8edc\n R10: 00007ffcf9df47e0 R11: 0000000000000246 R12: 0000000000671be0\n R13: 00007ffcf9df4e84 R14: 0000000000000008 R15: 0000000000000000\n\nFixes: f2cbd4852820 (\"net/sched: act_police: fix race condition on state variables\")\nReported-by: Cong Wang \u003cxiyou.wangcong@gmail.com\u003e\nSigned-off-by: Davide Caratti \u003cdcaratti@redhat.com\u003e\nAcked-by: Cong Wang \u003cxiyou.wangcong@gmail.com\u003e\nAcked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "605108acfe6233b72e2f803aa1cb59a2af3001ca", "tree": "7176a1fdfdd38c73fcf4d40ed76db3dcfc337b9e", "parents": [ "896585d48e8e9ba44cd1754fbce8537feffcc1a5" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Wed Nov 21 18:21:35 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 11:18:53 2018 -0800" }, "message": "net: don\u0027t keep lonely packets forever in the gro hash\n\nEric noted that with UDP GRO and NAPI timeout, we could keep a single\nUDP packet inside the GRO hash forever, if the related NAPI instance\ncalls napi_gro_complete() at an higher frequency than the NAPI timeout.\nWillem noted that even TCP packets could be trapped there, till the\nnext retransmission.\nThis patch tries to address the issue, flushing the old packets -\nthose with a NAPI_GRO_CB age before the current jiffy - before scheduling\nthe NAPI timeout. The rationale is that such a timeout should be\nwell below a jiffy and we are not flushing packets eligible for sane GRO.\n\nv1 -\u003e v2:\n - clarified the commit message and comment\n\nRFC -\u003e v1:\n - added \u0027Fixes tags\u0027, cleaned-up the wording.\n\nReported-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nFixes: 3b47d30396ba (\"net: gro: add a per device gro flush timer\")\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\nAcked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\nAcked-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "896585d48e8e9ba44cd1754fbce8537feffcc1a5", "tree": "bd98bbe64436a9df917e395b81b53feb01e301ef", "parents": [ "5cd8d46ea1562be80063f53c7c6a5f40224de623" ], "author": { "name": "Hangbin Liu", "email": "liuhangbin@gmail.com", "time": "Wed Nov 21 21:52:33 2018 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 11:17:56 2018 -0800" }, "message": "net/ipv6: re-do dad when interface has IFF_NOARP flag change\n\nWhen we add a new IPv6 address, we should also join corresponding solicited-node\nmulticast address, unless the interface has IFF_NOARP flag, as function\naddrconf_join_solict() did. But if we remove IFF_NOARP flag later, we do\nnot do dad and add the mcast address. So we will drop corresponding neighbour\ndiscovery message that came from other nodes.\n\nA typical example is after creating a ipvlan with mode l3, setting up an ipv6\naddress and changing the mode to l2. Then we will not be able to ping this\naddress as the interface doesn\u0027t join related solicited-node mcast address.\n\nFix it by re-doing dad when interface changed IFF_NOARP flag. Then we will add\ncorresponding mcast group and check if there is a duplicate address on the\nnetwork.\n\nReported-by: Jianlin Shi \u003cjishi@redhat.com\u003e\nReviewed-by: Stefano Brivio \u003csbrivio@redhat.com\u003e\nSigned-off-by: Hangbin Liu \u003cliuhangbin@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d88783b9c8849d88c3a75b7b9071cba072b47eba", "tree": "1c1ddeaa694bb336b10dececd2b669da9dc60cb3", "parents": [ "a03bac580ae743d5900af626ac63f7f8cd85def9", "829383e183728dec7ed9150b949cd6de64127809" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 11:15:27 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 11:15:27 2018 -0800" }, "message": "Merge tag \u0027iommu-fixes-v4.20-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu\n\nPull IOMMU fixes from Joerg Roedel:\n\n - Two fixes for the Intel VT-d driver to fix a NULL-ptr dereference and\n an unbalance in an allocate/free path (allocated with memremap, freed\n with iounmap)\n\n - Fix for a crash in the Renesas IOMMU driver\n\n - Fix for the Advanced Virtual Interrupt Controler (AVIC) code in the\n AMD IOMMU driver\n\n* tag \u0027iommu-fixes-v4.20-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:\n iommu/vt-d: Use memunmap to free memremap\n amd/iommu: Fix Guest Virtual APIC Log Tail Address Register\n iommu/ipmmu-vmsa: Fix crash on early domain free\n iommu/vt-d: Fix NULL pointer dereference in prq_event_thread()\n" }, { "commit": "5cd8d46ea1562be80063f53c7c6a5f40224de623", "tree": "53ce4843c87bbbeed12992b26dc4e15cf8470eaf", "parents": [ "039e70a70c8417b5bf5878a60612ebd2c95f731e" ], "author": { "name": "Willem de Bruijn", "email": "willemb@google.com", "time": "Tue Nov 20 13:00:18 2018 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Nov 23 11:08:03 2018 -0800" }, "message": "packet: copy user buffers before orphan or clone\n\ntpacket_snd sends packets with user pages linked into skb frags. It\nnotifies that pages can be reused when the skb is released by setting\nskb-\u003edestructor to tpacket_destruct_skb.\n\nThis can cause data corruption if the skb is orphaned (e.g., on\ntransmit through veth) or cloned (e.g., on mirror to another psock).\n\nCreate a kernel-private copy of data in these cases, same as tun/tap\nzerocopy transmission. Reuse that infrastructure: mark the skb as\nSKBTX_ZEROCOPY_FRAG, which will trigger copy in skb_orphan_frags(_rx).\n\nUnlike other zerocopy packets, do not set shinfo destructor_arg to\nstruct ubuf_info. tpacket_destruct_skb already uses that ptr to notify\nwhen the original skb is released and a timestamp is recorded. Do not\nchange this timestamp behavior. The ubuf_info-\u003ecallback is not needed\nanyway, as no zerocopy notification is expected.\n\nMark destructor_arg as not-a-uarg by setting the lower bit to 1. The\nresulting value is not a valid ubuf_info pointer, nor a valid\ntpacket_snd frame address. Add skb_zcopy_.._nouarg helpers for this.\n\nThe fix relies on features introduced in commit 52267790ef52 (\"sock:\nadd MSG_ZEROCOPY\"), so can be backported as is only to 4.14.\n\nTested with from `./in_netns.sh ./txring_overwrite` from\nhttp://github.com/wdebruij/kerneltools/tests\n\nFixes: 69e3c75f4d54 (\"net: TX_RING and packet mmap\")\nReported-by: Anand H. Krishnan \u003canandhkrishnan@gmail.com\u003e\nSigned-off-by: Willem de Bruijn \u003cwillemb@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a03bac580ae743d5900af626ac63f7f8cd85def9", "tree": "fc6a805dd06558b6e3327d4b2ac2d61df04cffe1", "parents": [ "b88af994872418f0a98db6f4a9bae849315a99b0", "2bbb5fa37475d7aa5fa62f34db1623f3da2dfdfa" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:56:16 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:56:16 2018 -0800" }, "message": "Merge tag \u0027acpi-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull ACPI fix from Rafael Wysocki:\n \"Prevent the ACPI core from registering a platform device for the\n SMB0001 HID to avoid IRQ allocation issues (Hans de Goede)\"\n\n* tag \u0027acpi-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n ACPI / platform: Add SMB0001 HID to forbidden_id_list\n" }, { "commit": "b88af994872418f0a98db6f4a9bae849315a99b0", "tree": "abbfed29d2affe081769ed63eef57cbe0c1bffa8", "parents": [ "e6005d3c42336074de3745718ac85807dd6e1e6a", "1d50088ca3956e5dcd2751a658e7869b9af10bb4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:52:57 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:52:57 2018 -0800" }, "message": "Merge tag \u0027pm-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management fixes from Rafael Wysocki:\n \"These fix two issues in the Operating Performance Points (OPP)\n framework, one cpufreq driver issue, one problem related to the tasks\n freezer and a few build-related issues in the cpupower utility.\n\n Specifics:\n\n - Fix tasks freezer deadlock in de_thread() that occurs if one of its\n sub-threads has been frozen already (Chanho Min).\n\n - Avoid registering a platform device by the ti-cpufreq driver on\n platforms that cannot use it (Dave Gerlach).\n\n - Fix a mistake in the ti-opp-supply operating performance points\n (OPP) driver that caused an incorrect reference voltage to be used\n and make it adjust the minimum voltage dynamically to avoid hangs\n or crashes in some cases (Keerthy).\n\n - Fix issues related to compiler flags in the cpupower utility and\n correct a linking problem in it by renaming a file with a duplicate\n name (Jiri Olsa, Konstantin Khlebnikov)\"\n\n* tag \u0027pm-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n exec: make de_thread() freezable\n cpufreq: ti-cpufreq: Only register platform_device when supported\n opp: ti-opp-supply: Correct the supply in _get_optimal_vdd_voltage call\n opp: ti-opp-supply: Dynamically update u_volt_min\n tools cpupower: Override CFLAGS assignments\n tools cpupower debug: Allow to use outside build flags\n tools/power/cpupower: fix compilation with STATIC\u003dtrue\n" }, { "commit": "4f9f49646a5733c0c2bd49940673dde89a9c5add", "tree": "8b97582d7df983e4a364554868a27b542e6c50d9", "parents": [ "b5d9a07ef7736b2456b9d3c90568de25e43d8ec3" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Wed Nov 21 15:07:00 2018 +0000" }, "committer": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Fri Nov 23 18:44:16 2018 +0000" }, "message": "arm64: cpufeature: Fix mismerge of CONFIG_ARM64_SSBD block\n\nWhen merging support for SSBD and the CRC32 instructions, the conflict\nresolution for the new capability entries in arm64_features[]\ninadvertedly predicated the availability of the CRC32 instructions on\nCONFIG_ARM64_SSBD, despite the functionality being entirely unrelated.\n\nMove the #ifdef CONFIG_ARM64_SSBD down so that it only covers the SSBD\ncapability.\n\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n" }, { "commit": "e6005d3c42336074de3745718ac85807dd6e1e6a", "tree": "eefc7be37bc396627330eb5896022010e5e2f948", "parents": [ "dcd3aa31dcdd6d8eae8d4771c44aeb3b1fec995a", "10547d956d590fbb846a7bd053a1dfc30e8f9a68" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:40:19 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:40:19 2018 -0800" }, "message": "Merge tag \u0027gpio-v4.20-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio\n\nPull GPIO fixes from Linus Walleij:\n \"Minor stuff except the IDA leak which was kind of important to fix.\n Also new maintainers, yay.\n\n - Do not lose an IDA on the gpiochip register errorpath.\n\n - Fix the PXA non-pincontrol GPIO-using platforms.\n\n - Fix the direction on the mockup GPIO driver.\n\n - Add some MAINTAINERS stuff: Bartosz stepped up as GPIO\n co-maintainer, and Andy established an Intel git tree\"\n\n* tag \u0027gpio-v4.20-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:\n MAINTAINERS: Do maintain Intel GPIO drivers via separate tree\n gpio: mockup: fix indicated direction\n gpio: pxa: fix legacy non pinctrl aware builds again\n gpio: don\u0027t free unallocated ida on gpiochip_add_data_with_key() error path\n MAINTAINERS: add myself as co-maintainer of gpiolib\n" }, { "commit": "dcd3aa31dcdd6d8eae8d4771c44aeb3b1fec995a", "tree": "fa1331a82b46ff86356c2350b101288fffc5a9f3", "parents": [ "9b7c880c834c0a1c80a1dc6b8a0b19155361321f", "5305ec6a27b2dc7398a689e661a4a2e951026f09" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:36:02 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:36:02 2018 -0800" }, "message": "Merge tag \u0027mmc-v4.20-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc\n\nPull MMC fixes from Ulf Hansson:\n \"MMC host:\n\n - sdhci-pci: Fixup card detect lookup\n\n - sdhci-pci: Workaround GLK firmware bug for tuning\"\n\n* tag \u0027mmc-v4.20-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:\n mmc: sdhci-pci: Workaround GLK firmware failing to restore the tuning value\n mmc: sdhci-pci: Try \"cd\" for card-detect lookup before using NULL\n" }, { "commit": "9b7c880c834c0a1c80a1dc6b8a0b19155361321f", "tree": "5885d049cf091568e128be5de1a25e988d7a938b", "parents": [ "edeca3a769ad28a9477798c3b1d8e0701db728e4", "98c9cdfd34fbb62886e4c5a07e33661aa3352ef5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:03:08 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Nov 23 10:03:08 2018 -0800" }, "message": "Merge tag \u0027drm-fixes-2018-11-23\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Regular drm fixes:\n\n amdgpu:\n - Vega20 fixes\n - firmware loading fix\n - panel display fix\n - override fix\n\n i915:\n - Sandybridge lockup fix\n - fastboot DSI panel fix\n - GPU hang on Broxton\n - GPU reloc fixes on pineview/bearlake\n\n ast:\n - screen blurring fix\n - cursor appearance fix\n\n udmabuf:\n - mmap fix\n\n vc4:\n - NULL deref fix\n - async cursor update fix\n\n All seems pretty normal at this stage\"\n\n* tag \u0027drm-fixes-2018-11-23\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/ast: fixed cursor may disappear sometimes\n drm/ast: change resolution may cause screen blurred\n drm/i915: Add rotation readout for plane initial config\n drm/i915: Force a LUT update in intel_initial_commit()\n drm/fb-helper: Blacklist writeback when adding connectors to fbdev\n drm/i915: Write GPU relocs harder with gen3\n drm/amdgpu: Enable HDP memory light sleep\n drm/i915: Prevent machine hang from Broxton\u0027s vtd w/a and error capture\n drm/amd/pp: handle negative values when reading OD\n drm/amdgpu: Add missing firmware entry for HAINAN\n drm/amd/powerplay: disable Vega20 DS related features\n drm/amdgpu: Fix oops when pp_funcs-\u003eswitch_power_profile is unset\n drm/i915: Disable LP3 watermarks on all SNB machines\n drm/ast: Remove existing framebuffers before loading driver\n udmabuf: set read/write flag when exporting\n drm/amd/display: Support amdgpu \"max bpc\" connector property (v2)\n drm/amdgpu: Add amdgpu \"max bpc\" connector property (v2)\n drm/vc4: Set -\u003elegacy_cursor_update to false when doing non-async updates\n drm/vc4: Fix NULL pointer dereference in the async update path\n" }, { "commit": "b5d9a07ef7736b2456b9d3c90568de25e43d8ec3", "tree": "30c045c7573b3374a022ff24cc70339ffc0b71f4", "parents": [ "9ff01193a20d391e8dbce4403dd5ef87c7eaaca6" ], "author": { "name": "Sergey Matyukevich", "email": "geomatsi@gmail.com", "time": "Fri Nov 16 21:21:30 2018 +0300" }, "committer": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Fri Nov 23 17:33:27 2018 +0000" }, "message": "arm64: sysreg: fix sparse warnings\n\nSpecify correct type for the constants to avoid\nthe following sparse complaints:\n\n./arch/arm64/include/asm/sysreg.h:471:42: warning: constant 0xffffffffffffffff is so big it is unsigned long\n./arch/arm64/include/asm/sysreg.h:512:42: warning: constant 0xffffffffffffffff is so big it is unsigned long\n\nAcked-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nAcked-by: Olof Johansson \u003colof@lixom.net\u003e\nAcked-by: Luc Van Oostenryck \u003cluc.vanoostenryck@gmail.com\u003e\nSigned-off-by: Sergey Matyukevich \u003cgeomatsi@gmail.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n" }, { "commit": "1d50088ca3956e5dcd2751a658e7869b9af10bb4", "tree": "5acc5b25f946cde7b2621e61780365bcd8489577", "parents": [ "bec00cb5e97c19d2c8bd10db15d334cb40760000", "d98ccfc3948ab63152494bb6b9c17e15295c0310", "c22397888f1eed98cd59f0a88f2a5f6925f80e15" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Nov 23 10:32:49 2018 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Nov 23 10:32:49 2018 +0100" }, "message": "Merge branches \u0027pm-cpufreq\u0027 and \u0027pm-sleep\u0027\n\n* pm-cpufreq:\n cpufreq: ti-cpufreq: Only register platform_device when supported\n\n* pm-sleep:\n exec: make de_thread() freezable\n" }, { "commit": "bec00cb5e97c19d2c8bd10db15d334cb40760000", "tree": "bf6dfafe9390855b024109f5a6b433e3d2fb52e5", "parents": [ "6eb5c9b5edffe626d2df6a22f449dc7fab253841", "0db699f747cc24835af77a34852882a33b3e1fc6" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Nov 23 10:32:22 2018 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Nov 23 10:32:22 2018 +0100" }, "message": "Merge branches \u0027pm-opp\u0027 and \u0027pm-tools\u0027\n\n* pm-opp:\n opp: ti-opp-supply: Correct the supply in _get_optimal_vdd_voltage call\n opp: ti-opp-supply: Dynamically update u_volt_min\n\n* pm-tools:\n tools cpupower: Override CFLAGS assignments\n tools cpupower debug: Allow to use outside build flags\n tools/power/cpupower: fix compilation with STATIC\u003dtrue\n" }, { "commit": "98c9cdfd34fbb62886e4c5a07e33661aa3352ef5", "tree": "053ad46392d3d150f9dfb91632d57cffe799420a", "parents": [ "8cf6f361eb76bf7fca85bde15a0a9316fa124c0c", "f559156c399cfb11d53a128d210118fbea36816e" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Nov 23 11:03:20 2018 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Nov 23 11:03:21 2018 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2018-11-22\u0027 of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes\n\n- Fix for fastboot DSI panel boot time flicker regression, also fixes Bugzilla #108225\n- Fix Bugzilla #101269 to avoid GPU hangs on Sandybridge machines\n- Avoid GPU hang on error capture on Broxton with Vt-d enabled\n- Avoid missing GPU relocations on Pineview and Bearlake (Gen3)\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20181122120555.GA18282@jlahtine-desk.ger.corp.intel.com\n" }, { "commit": "039e70a70c8417b5bf5878a60612ebd2c95f731e", "tree": "fcd3293be5b650116d47c5f758a64acb75e2e625", "parents": [ "6d0f60b0f8588fd4380ea5df9601e12fddd55ce2", "5bf032ef08e6a110edc1e3bfb3c66a208fb55125" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 22 11:53:26 2018 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 22 11:53:26 2018 -0800" }, "message": "Merge branch \u0027ibmvnic-Fix-queue-and-buffer-accounting-errors\u0027\n\nThomas Falcon says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nibmvnic: Fix queue and buffer accounting errors\n\nThis series includes two small fixes. The first resolves a typo bug\nin the code to clean up unused RX buffers during device queue removal.\nThe second ensures that device queue memory is updated to reflect new\nsupported queue ring sizes after migration to other backing hardware.\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": "5bf032ef08e6a110edc1e3bfb3c66a208fb55125", "tree": "fcd3293be5b650116d47c5f758a64acb75e2e625", "parents": [ "b7cdec3d699db2e5985ad39de0f25d3b6111928e" ], "author": { "name": "Thomas Falcon", "email": "tlfalcon@linux.ibm.com", "time": "Wed Nov 21 11:17:59 2018 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 22 11:53:25 2018 -0800" }, "message": "ibmvnic: Update driver queues after change in ring size support\n\nDuring device reset, queue memory is not being updated to accommodate\nchanges in ring buffer sizes supported by backing hardware. Track\nany differences in ring buffer sizes following the reset and update\nqueue memory when possible.\n\nSigned-off-by: Thomas Falcon \u003ctlfalcon@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b7cdec3d699db2e5985ad39de0f25d3b6111928e", "tree": "65db91dfa4cd7c6624b3ea74d079682fc4ce4eae", "parents": [ "6d0f60b0f8588fd4380ea5df9601e12fddd55ce2" ], "author": { "name": "Thomas Falcon", "email": "tlfalcon@linux.ibm.com", "time": "Wed Nov 21 11:17:58 2018 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 22 11:53:25 2018 -0800" }, "message": "ibmvnic: Fix RX queue buffer cleanup\n\nThe wrong index is used when cleaning up RX buffer objects during release\nof RX queues. Update to use the correct index counter.\n\nSigned-off-by: Thomas Falcon \u003ctlfalcon@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6d0f60b0f8588fd4380ea5df9601e12fddd55ce2", "tree": "333aeb0e376635b4c54c57b70908191e8a287af6", "parents": [ "0211dda68a4f6531923a2f72d8e8959207f59fba" ], "author": { "name": "Lorenzo Bianconi", "email": "lorenzo.bianconi@redhat.com", "time": "Wed Nov 21 16:32:10 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 22 11:40:30 2018 -0800" }, "message": "net: thunderx: set xdp_prog to NULL if bpf_prog_add fails\n\nSet xdp_prog pointer to NULL if bpf_prog_add fails since that routine\nreports the error code instead of NULL in case of failure and xdp_prog\npointer value is used in the driver to verify if XDP is currently\nenabled.\nMoreover report the error code to userspace if nicvf_xdp_setup fails\n\nFixes: 05c773f52b96 (\"net: thunderx: Add basic XDP support\")\nSigned-off-by: Lorenzo Bianconi \u003clorenzo.bianconi@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0211dda68a4f6531923a2f72d8e8959207f59fba", "tree": "1c50a0a281fd47b17ed6fe90a99577c51f6b9636", "parents": [ "426a593e641ebf0d9288f0a2fcab644a86820220" ], "author": { "name": "Tal Gilboa", "email": "talgi@mellanox.com", "time": "Wed Nov 21 16:28:23 2018 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Nov 22 11:36:54 2018 -0800" }, "message": "net/dim: Update DIM start sample after each DIM iteration\n\nOn every iteration of net_dim, the algorithm may choose to\ncheck for the system state by comparing current data sample\nwith previous data sample. After each of these comparison,\nregardless of the action taken, the sample used as baseline\nis needed to be updated.\n\nThis patch fixes a bug that causes DIM to take wrong decisions,\ndue to never updating the baseline sample for comparison between\niterations. This way, DIM always compares current sample with\nzeros.\n\nAlthough this is a functional fix, it also improves and stabilizes\nperformance as the algorithm works properly now.\n\nPerformance:\nTested single UDP TX stream with pktgen:\nsamples/pktgen/pktgen_sample03_burst_single_flow.sh -i p4p2 -d 1.1.1.1\n-m 24:8a:07:88:26:8b -f 3 -b 128\n\nConnectX-5 100GbE packet rate improved from 15-19Mpps to 19-20Mpps.\nAlso, toggling between profiles is less frequent with the fix.\n\nFixes: 8115b750dbcb (\"net/dim: use struct net_dim_sample as arg to net_dim\")\nSigned-off-by: Tal Gilboa \u003ctalgi@mellanox.com\u003e\nReviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bb21ce0ad227b69ec0f83279297ee44232105d96", "tree": "2e90fa3820d7c3e2642fae6f617f131115052fc6", "parents": [ "99f2c55591fb5c1b536263970d98c2ebc2089906" ], "author": { "name": "Tigran Mkrtchyan", "email": "tigran.mkrtchyan@desy.de", "time": "Wed Nov 21 12:25:41 2018 +0100" }, "committer": { "name": "Trond Myklebust", "email": "trond.myklebust@hammerspace.com", "time": "Thu Nov 22 14:04:55 2018 -0500" }, "message": "flexfiles: use per-mirror specified stateid for IO\n\nrfc8435 says:\n\n For tight coupling, ffds_stateid provides the stateid to be used by\n the client to access the file.\n\nHowever current implementation replaces per-mirror provided stateid with\nby open or lock stateid.\n\nEnsure that per-mirror stateid is used by ff_layout_write_prepare_v4 and\nnfs4_ff_layout_prepare_ds.\n\nSigned-off-by: Tigran Mkrtchyan \u003ctigran.mkrtchyan@desy.de\u003e\nSigned-off-by: Rick Macklem \u003crmacklem@uoguelph.ca\u003e\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n" }, { "commit": "99f2c55591fb5c1b536263970d98c2ebc2089906", "tree": "3bb865138830f2d44759f49073e61250125d787b", "parents": [ "aeabb3c96186a0f944fc2b1f25c84d5eb3a93fa9" ], "author": { "name": "Olga Kornievskaia", "email": "kolga@netapp.com", "time": "Wed Nov 21 11:24:22 2018 -0500" }, "committer": { "name": "Trond Myklebust", "email": "trond.myklebust@hammerspace.com", "time": "Thu Nov 22 13:54:46 2018 -0500" }, "message": "NFSv4.2 copy do not allocate memory under the lock\n\nBruce pointed out that we shouldn\u0027t allocate memory while holding\na lock in the nfs4_callback_offload() and handle_async_copy()\nthat deal with a racing CB_OFFLOAD and reply to COPY case.\n\nSigned-off-by: Olga Kornievskaia \u003ckolga@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n" }, { "commit": "edeca3a769ad28a9477798c3b1d8e0701db728e4", "tree": "c410b77f2418fc561ec46da8f25fc653abbb46db", "parents": [ "52465bce85a2d28bcec5cba5a645bb610367ab1b", "a6b0961b39896a9f9f1350d26d202f078a7d9dbc" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:45:44 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:45:44 2018 -0800" }, "message": "Merge tag \u0027sound-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound\n\nPull sound fixes from Takashi Iwai:\n \"The only significant change is for OSS PCM emulation to convert with\n kvcalloc() to address both performance and security issues. It\u0027s a\n pretty straightforward change, which should be safe.\n\n The rest are, as usual, device-specific small fixes for HD-audio\"\n\n* tag \u0027sound-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:\n ALSA: hda/ca0132 - fix AE-5 pincfg\n ALSA: hda/ca0132 - Add new ZxR quirk\n ALSA: hda/ca0132 - Call pci_iounmap() instead of iounmap()\n ALSA: hda/realtek - Add quirk entry for HP Pavilion 15\n ALSA: oss: Use kvzalloc() for local buffer allocations\n" }, { "commit": "52465bce85a2d28bcec5cba5a645bb610367ab1b", "tree": "90d8c46136af41d90f3dceae2a01f1f70e158855", "parents": [ "4cd731953d620b7e4e999a90d13db58b88c5e95b", "544b03da39e2d7b4961d3163976ed4bfb1fac509" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:43:06 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:43:06 2018 -0800" }, "message": "Merge tag \u0027char-misc-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc\n\nPull char/misc driver fixes from Greg KH:\n \"Here are some small char/misc driver fixes for issues that have been\n reported.\n\n Nothing major, highlights include:\n\n - gnss sync write fixes\n\n - uio oops fix\n\n - nvmem fixes\n\n - other minor fixes and some documentation/maintainers updates\n\n Full details are in the shortlog.\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027char-misc-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:\n Documentation/security-bugs: Postpone fix publication in exceptional cases\n MAINTAINERS: Add Sasha as a stable branch maintainer\n gnss: sirf: fix synchronous write timeout\n gnss: serial: fix synchronous write timeout\n uio: Fix an Oops on load\n test_firmware: fix error return getting clobbered\n nvmem: core: fix regression in of_nvmem_cell_get()\n misc: atmel-ssc: Fix section annotation on atmel_ssc_get_driver_data\n drivers/misc/sgi-gru: fix Spectre v1 vulnerability\n Drivers: hv: kvp: Fix the recent regression caused by incorrect clean-up\n slimbus: ngd: remove unnecessary check\n" }, { "commit": "4cd731953d620b7e4e999a90d13db58b88c5e95b", "tree": "5ffa6ac38f4dc0d60a6e811eaef2c0d2f8385e69", "parents": [ "ef4d6f2c0c659922856bb48cbb7a83ac97941e01", "63529eaa6164ef7ab4b907b25ac3648177e5e78f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:39:29 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:39:29 2018 -0800" }, "message": "Merge tag \u0027usb-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nPull USB fixes from Greg KH:\n \"Here are a number of small USB fixes for 4.20-rc4.\n\n There\u0027s the usual xhci and dwc2/3 fixes as well as a few minor other\n issues resolved for problems that have been reported. Full details are\n in the shortlog.\n\n All have been in linux-next for a while with no reported issues\"\n\n* tag \u0027usb-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:\n usb: cdc-acm: add entry for Hiro (Conexant) modem\n usb: xhci: Prevent bus suspend if a port connect change or polling state is detected\n usb: core: Fix hub port connection events lost\n usb: dwc3: gadget: fix ISOC TRB type on unaligned transfers\n Revert \"usb: gadget: ffs: Fix BUG when userland exits with submitted AIO transfers\"\n usb: dwc2: pci: Fix an error code in probe\n usb: dwc3: Fix NULL pointer exception in dwc3_pci_remove()\n xhci: Add quirk to workaround the errata seen on Cavium Thunder-X2 Soc\n usb: xhci: fix timeout for transition from RExit to U0\n usb: xhci: fix uninitialized completion when USB3 port got wrong status\n xhci: Add check for invalid byte size error when UAS devices are connected.\n xhci: handle port status events for removed USB3 hcd\n xhci: Fix leaking USB3 shared_hcd at xhci removal\n USB: misc: appledisplay: add 20\" Apple Cinema Display\n USB: quirks: Add no-lpm quirk for Raydium touchscreens\n usb: quirks: Add delay-init quirk for Corsair K70 LUX RGB\n USB: Wait for extra delay time after USB_PORT_FEAT_RESET for quirky hub\n usb: dwc3: gadget: Properly check last unaligned/zero chain TRB\n usb: dwc3: core: Clean up ULPI device\n" }, { "commit": "ef4d6f2c0c659922856bb48cbb7a83ac97941e01", "tree": "1af9ccde71590d0920c68497100cf0c386b6138c", "parents": [ "b84b6345e3827ab616946b52f46e95179657b596", "e8828ec1c003727fc001eab06aa19bd2ca9b677e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:35:30 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:35:30 2018 -0800" }, "message": "Merge tag \u0027mtd/fixes-for-4.20-rc4\u0027 of git://git.infradead.org/linux-mtd\n\nPull mtd fixes from Boris Brezillon:\n \"SPI NOR fixes:\n\n - Various fixes related to the SFDP parsing code merged in 4.20\n\n - Fix for a page fault in the cadence-qspi\n\n NAND fixes:\n\n - Fix a macro name conflict between the QCOM NAND controller driver\n and the RISC-V asm headers\n\n - Fix of-node handling in the atmel driver\"\n\n* tag \u0027mtd/fixes-for-4.20-rc4\u0027 of git://git.infradead.org/linux-mtd:\n mtd: spi-nor: fix selection of uniform erase type in flexible conf\n mtd: spi-nor: Fix Cadence QSPI page fault kernel panic\n mtd: rawnand: qcom: Namespace prefix some commands\n mtd: rawnand: atmel: fix OF child-node lookup\n mtd: spi_nor: pass DMA-able buffer to spi_nor_read_raw()\n mtd: spi-nor: don\u0027t overwrite errno in spi_nor_get_map_in_use()\n mtd: spi-nor: fix iteration over smpt array\n mtd: spi-nor: don\u0027t drop sfdp data if optional parsers fail\n" }, { "commit": "b84b6345e3827ab616946b52f46e95179657b596", "tree": "db3dfb97bb824f3206ae552d0854b49e47186515", "parents": [ "92b419289ceecdd1eae03114928913f298b84327", "8e4829c6f7470722c1f5a1dc5769ebe09ef036d6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:31:46 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 22 08:31:46 2018 -0800" }, "message": "Merge tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi\n\nPull SCSI fixes from James Bottomley:\n \"Two small fixes.\n\n The qla2xxx is a regression from 4.18 and the ufs one is a device\n enablement fix\"\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n scsi: ufs: Fix hynix ufs bug with quirk on hi36xx SoC\n scsi: qla2xxx: Timeouts occur on surprise removal of QLogic adapter\n" }, { "commit": "829383e183728dec7ed9150b949cd6de64127809", "tree": "3bfcd7b3832449e9c15588987664a642d961ad58", "parents": [ "ab99be4683d9db33b100497d463274ebd23bd67e" ], "author": { "name": "Pan Bian", "email": "bianpan2016@163.com", "time": "Wed Nov 21 17:53:47 2018 +0800" }, "committer": { "name": "Joerg Roedel", "email": "jroedel@suse.de", "time": "Thu Nov 22 17:02:21 2018 +0100" }, "message": "iommu/vt-d: Use memunmap to free memremap\n\nmemunmap() should be used to free the return of memremap(), not\niounmap().\n\nFixes: dfddb969edf0 (\u0027iommu/vt-d: Switch from ioremap_cache to memremap\u0027)\nSigned-off-by: Pan Bian \u003cbianpan2016@163.com\u003e\nSigned-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n" }, { "commit": "ffe0e7cf290f5c9d1392134b4ef8da2a3761a4cd", "tree": "c214b24f1c92a03f42bdacf14c03bf270f709409", "parents": [ "f1539a0c2545d7bd82e451bd1464f2a820f55de4" ], "author": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Wed Nov 21 16:27:12 2018 +0100" }, "committer": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Thu Nov 22 08:57:44 2018 +0100" }, "message": "Revert \"Input: Add the `REL_WHEEL_HI_RES` event code\"\n\nThis reverts commit aaf9978c3c0291ef3beaa97610bc9c3084656a85.\n\nQuoting Peter:\n\nThere is a HID feature report called \"Resolution Multiplier\"\nDescribed in the \"Enhanced Wheel Support in Windows\" doc and\nthe \"USB HID Usage Tables\" page 30.\n\nhttp://download.microsoft.com/download/b/d/1/bd1f7ef4-7d72-419e-bc5c-9f79ad7bb66e/wheel.docx\nhttps://www.usb.org/sites/default/files/documents/hut1_12v2.pdf\n\nThis was new for Windows Vista, so we\u0027re only a decade behind here. I only\naccidentally found this a few days ago while debugging a stuck button on a\nMicrosoft mouse.\n\nThe docs above describe it like this: a wheel control by default sends\nvalue 1 per notch. If the resolution multiplier is active, the wheel is\nexpected to send a value of $multiplier per notch (e.g. MS Sculpt mouse) or\njust send events more often, i.e. for less physical motion (e.g. MS Comfort\nmouse).\n\nFor the latter, you need the right HW of course. The Sculpt mouse has\ntactile wheel clicks, so nothing really changes. The Comfort mouse has\ncontinuous motion with no tactile clicks. Similar to the free-wheeling\nLogitech mice but without any inertia.\n\nNote that the doc also says that Vista and onwards *always* enable this\nfeature where available.\n\nAn example HID definition looks like this:\n\n Usage Page Generic Desktop (0x01)\n Usage Resolution Multiplier (0x48)\n Logical Minimum 0\n Logical Maximum 1\n Physical Minimum 1\n Physical Maximum 16\n Report Size 2 # in bits\n Report Count 1\n Feature (Data, Var, Abs)\n\nSo the actual bits have values 0 or 1 and that reflects real values 1 or 16.\nWe\u0027ve only seen single-bits so far, so there\u0027s low-res and hi-res, but\nnothing in between.\n\nThe multiplier is available for HID usages \"Wheel\" and \"AC Pan\" (horiz wheel).\nMicrosoft suggests that\n\n\u003e Vendors should ship their devices with smooth scrolling disabled and allow\n\u003e Windows to enable it. This ensures that the device works like a regular HID\n\u003e device on legacy operating systems that do not support smooth scrolling.\n(see the wheel doc linked above)\n\nThe mice that we tested so far do reset on unplug.\n\nDevice Support looks to be all (?) Microsoft mice but nothing else\n\nNot supported:\n- Logitech G500s, G303\n- Roccat Kone XTD\n- all the cheap Lenovo, HP, Dell, Logitech USB mice that come with a\n workstation that I could find don\u0027t have it.\n- Etekcity something something\n- Razer Imperator\n\nSupported:\n- Microsoft Comfort Optical Mouse 3000 - yes, physical: 1:4\n- Microsoft Sculpt Ergonomic Mouse - yes, physical: 1:12\n- Microsoft Surface mouse - yes, physical: 1:4\n\nSo again, I think this is really just available on Microsoft mice, but\nprobably all decent MS mice released over the last decade.\n\nLooking at the hardware itself:\n\n- no noticeable notches in the weel\n- low-res: 18 events per 360deg rotation (click angle 20 deg)\n- high-res: 72 events per 360deg → matches multiplier of 4\n\n- I can feel the notches during wheel turns\n- low-res: 24 events per 360 deg rotation (click angle 15 deg)\n - horiz wheel is tilt-based, continuous output value 1\n- high-res: 24 events per 360deg with value 12 → matches multiplier of 12\n - horiz wheel output rate doubles/triples?, values is 3\n\n- It\u0027s a touch strip, not a wheel so no notches\n- high-res: events have value 4 instead of 1\n a bit strange given that it doesn\u0027t actually have notches.\n\nOk, why is this an issue for the current API? First, because the logitech\nmultiplier used in Harry\u0027s patches looks suspiciously like the Resolution\nMultiplier so I think we should assume it\u0027s the same thing. Nestor, can you\nshed some light on that?\n\n- `REL_WHEEL` is defined as the number of notches, emulated where needed.\n- `REL_WHEEL_HI_RES` is the movement of the user\u0027s finger in microns.\n- `WM_MOUSEWHEEL` (Windows) is is a multiple of 120, defined as \"the threshold\n for action to be taken and one such action\"\n https://docs.microsoft.com/en-us/windows/desktop/inputdev/wm-mousewheel\n\nIf the multiplier is set to M, this means we need an accumulated value of M\nuntil we can claim there was a wheel click. So after enabling the multiplier\nand setting it to the maximum (like Windows):\n- M units are 15deg rotation → 1 unit is 2620/M micron (see below). This is\n the `REL_WHEEL_HI_RES` value.\n - wheel diameter 20mm: 15 deg rotation is 2.62mm, 2620 micron (pi * 20mm /\n (360deg/15deg))\n- For every M units accumulated, send one `REL_WHEEL` event\n\nThe problem here is that we\u0027ve now hardcoded 20mm/15 deg into the kernel and\nwe have no way of getting the size of the wheel or the click angle into the\nkernel.\n\nIn userspace we now have to undo the kernel\u0027s calculation. If our click angle\nis e.g. 20 degree we have to undo the (lossy) calculation from the kernel and\ncalculate the correct angle instead. This also means the 15 is a hardcoded\noption forever and cannot be changed.\n\nIn hid-logitech-hidpp.c, the microns per unit is hardcoded per device.\nHarry, did you measure those by hand? We\u0027d need to update the kernel for\nevery device and there are 10 years worth of devices from MS alone.\n\nThe multiplier default is 8 which is in the right ballpark, so I\u0027m pretty\nsure this is the same as the Resolution Multiplier, just in HID++ lingo. And\ngiven that the 120 magic factor is what Windows uses in the end, I can\u0027t\nimagine Logitech rolling their own thing here. Nestor?\n\nAnd we\u0027re already fairly inaccurate with the microns anyway. The MX Anywhere\n2S has a click angle of 20 degrees (18 stops) and a 17mm wheel, so a wheel\nnotch is approximately 2.67mm, one event at multiplier 8 (1/8 of a notch)\nwould be 334 micron. That\u0027s only 80% of the fallback value of 406 in the\nkernel. Multiplier 6 gives us 445micron (10% off). I\u0027m assuming multiplier 7\ndoesn\u0027t exist because it\u0027s not a factor of 120.\n\nSummary:\n\nBest option may be to simply do what Windows is doing, all the HW manufacturers\nhave to use that approach after all. Switch `REL_WHEEL_HI_RES` to report in\nfractions of 120, with 120 being one notch and divide that by the multiplier\nfor the actual events. So e.g. the Logitech multiplier 8 would send value 15\nfor each event in hi-res mode. This can be converted in userspace to\nwhatever userspace needs (combined with a hwdb there that tells you wheel\nsize/click angle/...).\n\nConflicts:\n\tinclude/uapi/linux/input-event-codes.h -\u003e I kept the new\n reserved event in the code, so I had to adapt the revert\n slightly\n\nSigned-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nAcked-by: Harry Cutts \u003chcutts@chromium.org\u003e\nAcked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nAcked-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "f1539a0c2545d7bd82e451bd1464f2a820f55de4", "tree": "a1e53314077b8112e8b2d7224e1c784c5545fe54", "parents": [ "00acc9e2c4a55045d73f2615429742369996d801" ], "author": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Wed Nov 21 16:27:11 2018 +0100" }, "committer": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Thu Nov 22 08:57:38 2018 +0100" }, "message": "Revert \"HID: input: Create a utility class for counting scroll events\"\n\nThis reverts commit 1ff2e1a44e02d4bdbb9be67c7d9acc240a67141f.\n\nIt turns out the current API is not that compatible with\nsome Microsoft mice, so better start again from scratch.\n\nSigned-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nAcked-by: Harry Cutts \u003chcutts@chromium.org\u003e\nAcked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nAcked-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "00acc9e2c4a55045d73f2615429742369996d801", "tree": "78fe59327becd57016bc7e41c4994ab171a233a4", "parents": [ "5372fc3797691065b8318f0cc9ab8695cebb17e1" ], "author": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Wed Nov 21 16:27:10 2018 +0100" }, "committer": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Thu Nov 22 08:57:33 2018 +0100" }, "message": "Revert \"HID: logitech: Add function to enable HID++ 1.0 \"scrolling acceleration\"\"\n\nThis reverts commit 051dc9b0579602bd63e9df74d0879b5293e71581.\n\nIt turns out the current API is not that compatible with\nsome Microsoft mice, so better start again from scratch.\n\nSigned-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nAcked-by: Harry Cutts \u003chcutts@chromium.org\u003e\nAcked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nAcked-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "5372fc3797691065b8318f0cc9ab8695cebb17e1", "tree": "e52b24cf3261ae6cdc9ced7a58f20b0e340b48b5", "parents": [ "a69616d5b3e13f18cdec1c759818aec87866ac0c" ], "author": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Wed Nov 21 16:27:09 2018 +0100" }, "committer": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Thu Nov 22 08:57:27 2018 +0100" }, "message": "Revert \"HID: logitech: Enable high-resolution scrolling on Logitech mice\"\n\nThis reverts commit d56ca9855bf924f3bc9807a3e42f38539df3f41f.\n\nIt turns out the current API is not that compatible with\nsome Microsoft mice, so better start again from scratch.\n\nSigned-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nAcked-by: Harry Cutts \u003chcutts@chromium.org\u003e\nAcked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nAcked-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "a69616d5b3e13f18cdec1c759818aec87866ac0c", "tree": "97dbd5b5d299c571af51e8de4f518bc1114f8205", "parents": [ "d0341efe7449860e1bbfdd282b3b7f0c34309222" ], "author": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Wed Nov 21 16:27:08 2018 +0100" }, "committer": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Thu Nov 22 08:57:21 2018 +0100" }, "message": "Revert \"HID: logitech: Use LDJ_DEVICE macro for existing Logitech mice\"\n\nThis reverts commit 3fe1d6bbcd16f384d2c7dab2caf8e4b2df9ea7e6.\n\nIt turns out the current API is not that compatible with\nsome Microsoft mice, so better start again from scratch.\n\nSigned-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nAcked-by: Harry Cutts \u003chcutts@chromium.org\u003e\nAcked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nAcked-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "d0341efe7449860e1bbfdd282b3b7f0c34309222", "tree": "e4f25e977625c51ac0e79a4816d30ef865d854b2", "parents": [ "e2b95b27196cc90f37016d79432fb2af3f9c9291" ], "author": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Wed Nov 21 16:27:07 2018 +0100" }, "committer": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Thu Nov 22 08:57:03 2018 +0100" }, "message": "Revert \"HID: logitech: fix a used uninitialized GCC warning\"\n\nThis reverts commit 5fe2ccbef9d7aecf5c4402c753444f1a12096cfd.\n\nIt turns out the current API is not that compatible with\nsome Microsoft mice, so better start again from scratch.\n\nSigned-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nAcked-by: Harry Cutts \u003chcutts@chromium.org\u003e\nAcked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nAcked-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "e2b95b27196cc90f37016d79432fb2af3f9c9291", "tree": "3dc28e0204a9fd60f95035458b2b07f923e444e0", "parents": [ "fb862c3b199d28bee238d52e8270eae8650d6cb0" ], "author": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Wed Nov 21 16:27:06 2018 +0100" }, "committer": { "name": "Benjamin Tissoires", "email": "benjamin.tissoires@redhat.com", "time": "Thu Nov 22 08:56:04 2018 +0100" }, "message": "Revert \"HID: input: simplify/fix high-res scroll event handling\"\n\nThis reverts commit 044ee890286153a1aefb40cb8b6659921aecb38b.\n\nIt turns out the current API is not that compatible with\nsome Microsoft mice, so better start again from scratch.\n\nSigned-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nAcked-by: Harry Cutts \u003chcutts@chromium.org\u003e\nAcked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nAcked-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "8cf6f361eb76bf7fca85bde15a0a9316fa124c0c", "tree": "508bd62e730c0ba0b58df0c36f082e62aeb89c31", "parents": [ "1d74f133252fab157b9bcd4c710cee7ffcdd01c1", "a5d0f4565996e5595a10cb57b3d1e3d74379c502" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 22 11:19:20 2018 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 22 11:45:37 2018 +1000" }, "message": "Merge branch \u0027drm-fixes-4.20\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-fixes\n\n- OD fixes for powerplay\n- Vega20 fixes\n- KFD fix for Kaveri\n- add missing firmware declaration for hainan (SI chip)\n- Fix DC user experience regressions related to panels that support \u003e8 bpc\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Alex Deucher \u003calexdeucher@gmail.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20181121163647.2847-1-alexander.deucher@amd.com\n" }, { "commit": "1d74f133252fab157b9bcd4c710cee7ffcdd01c1", "tree": "7055fd74231e7c51b15c12c54281a41e168b0558", "parents": [ "7989b9ee8bafe5cc625381dd0c3c4586de27ca26", "8fd3b90300bec541806dac271de2fd44e2e4e2d2" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 22 11:17:46 2018 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 22 11:18:00 2018 +1000" }, "message": "Merge tag \u0027drm-misc-fixes-2018-11-21\u0027 of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes\n\n- vc4: Fix NULL deref in async path (Boris)\n- vc4: Avoid taking async path for cursor updates when impossible (Boris)\n- udmabuf: Fix mmap with PROT_WRITE (Gerd)\n- fb-helper: Don\u0027t use writeback connectors for fbdev (Paul)\n\nCc: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nCc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nCc: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Sean Paul \u003csean@poorly.run\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20181121155248.GA241511@art_vandelay\n" }, { "commit": "7989b9ee8bafe5cc625381dd0c3c4586de27ca26", "tree": "514cbd659488f297fc23ad2edd985dccda4c25eb", "parents": [ "1a37bd823891568f8721989aed0615835632d81a" ], "author": { "name": "Y.C. Chen", "email": "yc_chen@aspeedtech.com", "time": "Tue Oct 30 11:34:46 2018 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 22 11:16:02 2018 +1000" }, "message": "drm/ast: fixed cursor may disappear sometimes\n\nSigned-off-by: Y.C. Chen \u003cyc_chen@aspeedtech.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nReviewed-by: Dave Airlie \u003cairlied@redhat.com\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "426a593e641ebf0d9288f0a2fcab644a86820220", "tree": "9da1b8b14386d93e424e6e361567697d2f4bf392", "parents": [ "395048ebd4547d3cced71807b8aabfbde000fe55" ], "author": { "name": "Vincent Chen", "email": "vincentc@andestech.com", "time": "Wed Nov 21 09:38:11 2018 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 16:57:39 2018 -0800" }, "message": "net: faraday: ftmac100: remove netif_running(netdev) check before disabling interrupts\n\nIn the original ftmac100_interrupt(), the interrupts are only disabled when\nthe condition \"netif_running(netdev)\" is true. However, this condition\ncauses kerenl hang in the following case. When the user requests to\ndisable the network device, kernel will clear the bit __LINK_STATE_START\nfrom the dev-\u003estate and then call the driver\u0027s ndo_stop function. Network\ndevice interrupts are not blocked during this process. If an interrupt\noccurs between clearing __LINK_STATE_START and stopping network device,\nkernel cannot disable the interrupts due to the condition\n\"netif_running(netdev)\" in the ISR. Hence, kernel will hang due to the\ncontinuous interruption of the network device.\n\nIn order to solve the above problem, the interrupts of the network device\nshould always be disabled in the ISR without being restricted by the\ncondition \"netif_running(netdev)\".\n\n[V2]\nRemove unnecessary curly braces.\n\nSigned-off-by: Vincent Chen \u003cvincentc@andestech.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1a37bd823891568f8721989aed0615835632d81a", "tree": "5a917b4bab660b949d8872f9b46e4331d68faf2a", "parents": [ "5478ad10e7850ce3d8b7056db05ddfa3c9ddad9a" ], "author": { "name": "Y.C. Chen", "email": "yc_chen@aspeedtech.com", "time": "Wed Oct 03 14:57:47 2018 +0800" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 22 10:56:17 2018 +1000" }, "message": "drm/ast: change resolution may cause screen blurred\n\nThe value of pitches is not correct while calling mode_set.\nThe issue we found so far on following system:\n- Debian8 with XFCE Desktop\n- Ubuntu with KDE Desktop\n- SUSE15 with KDE Desktop\n\nSigned-off-by: Y.C. Chen \u003cyc_chen@aspeedtech.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nTested-by: Jean Delvare \u003cjdelvare@suse.de\u003e\nReviewed-by: Jean Delvare \u003cjdelvare@suse.de\u003e\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n" }, { "commit": "395048ebd4547d3cced71807b8aabfbde000fe55", "tree": "7325b43e2c0690f4f0944ae97096fda4cadefade", "parents": [ "1e2b1046b53f356e663fd145d0c57bea04a0343a", "e438bae43c1e08e688c09c410407b59fc1c173b4" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 16:14:56 2018 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 16:14:56 2018 -0800" }, "message": "Merge branch \u0027smc-fixes\u0027\n\nUrsula Braun says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet/smc: fixes 2018-11-12\n\nhere is V4 of some net/smc fixes in different areas for the net tree.\n\nv1-\u003ev2:\n do not define 8-byte alignment for union smcd_cdc_cursor in\n patch 4/5 \"net/smc: atomic SMCD cursor handling\"\nv2-\u003ev3:\n stay with 8-byte alignment for union smcd_cdc_cursor in\n patch 4/5 \"net/smc: atomic SMCD cursor handling\", but get rid of\n __packed for struct smcd_cdc_msg\nv3-\u003ev4:\n get rid of another __packed for struct smc_cdc_msg in\n patch 4/5 \"net/smc: atomic SMCD cursor handling\"\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": "e438bae43c1e08e688c09c410407b59fc1c173b4", "tree": "7325b43e2c0690f4f0944ae97096fda4cadefade", "parents": [ "b9a22dd9811dbcddb5623c499e5b736400059df6" ], "author": { "name": "Ursula Braun", "email": "ursula.braun@linux.ibm.com", "time": "Tue Nov 20 16:46:43 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 16:14:56 2018 -0800" }, "message": "net/smc: use after free fix in smc_wr_tx_put_slot()\n\nIn smc_wr_tx_put_slot() field pend-\u003eidx is used after being\ncleared. That means always idx 0 is cleared in the wr_tx_mask.\nThis results in a broken administration of available WR send\npayload buffers.\n\nSigned-off-by: Ursula Braun \u003cubraun@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b9a22dd9811dbcddb5623c499e5b736400059df6", "tree": "aa6958e351f6bee0593326693fca2878ae14c6e6", "parents": [ "0512f69e388c963dbe955d4bd9ae0f7d88d2dc54" ], "author": { "name": "Ursula Braun", "email": "ubraun@linux.ibm.com", "time": "Tue Nov 20 16:46:42 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 16:14:56 2018 -0800" }, "message": "net/smc: atomic SMCD cursor handling\n\nRunning uperf tests with SMCD on LPARs results in corrupted cursors.\nSMCD cursors should be treated atomically to fix cursor corruption.\n\nSigned-off-by: Ursula Braun \u003cubraun@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0512f69e388c963dbe955d4bd9ae0f7d88d2dc54", "tree": "38627f33d9ef54a2c83beb2628ec2bbf7c663827", "parents": [ "ee05ff7af26509f39360534a5225ee714416cdfd" ], "author": { "name": "Hans Wippel", "email": "hwippel@linux.ibm.com", "time": "Tue Nov 20 16:46:41 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 16:14:56 2018 -0800" }, "message": "net/smc: add SMC-D shutdown signal\n\nWhen a SMC-D link group is freed, a shutdown signal should be sent to\nthe peer to indicate that the link group is invalid. This patch adds the\nshutdown signal to the SMC code.\n\nSigned-off-by: Hans Wippel \u003chwippel@linux.ibm.com\u003e\nSigned-off-by: Ursula Braun \u003cubraun@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ee05ff7af26509f39360534a5225ee714416cdfd", "tree": "986949c0bf44e3269e288c385759b4a516183059", "parents": [ "f07920ad9c6f5781c90ac4915f8254d999d8c1cc" ], "author": { "name": "Karsten Graul", "email": "kgraul@linux.ibm.com", "time": "Tue Nov 20 16:46:40 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 16:14:56 2018 -0800" }, "message": "net/smc: use queue pair number when matching link group\n\nWhen searching for an existing link group the queue pair number is also\nto be taken into consideration. When the SMC server sends a new number\nin a CLC packet (keeping all other values equal) then a new link group\nis to be created on the SMC client side.\n\nSigned-off-by: Karsten Graul \u003ckgraul@linux.ibm.com\u003e\nSigned-off-by: Ursula Braun \u003cubraun@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f07920ad9c6f5781c90ac4915f8254d999d8c1cc", "tree": "73c7617e7b78bf570c380a7041b6d1916e02b68e", "parents": [ "1e2b1046b53f356e663fd145d0c57bea04a0343a" ], "author": { "name": "Hans Wippel", "email": "hwippel@linux.ibm.com", "time": "Tue Nov 20 16:46:39 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 16:14:56 2018 -0800" }, "message": "net/smc: abort CLC connection in smc_release\n\nIn case of a non-blocking SMC socket, the initial CLC handshake is\nperformed over a blocking TCP connection in a worker. If the SMC socket\nis released, smc_release has to wait for the blocking CLC socket\noperations (e.g., kernel_connect) inside the worker.\n\nThis patch aborts a CLC connection when the respective non-blocking SMC\nsocket is released to avoid waiting on socket operations or timeouts.\n\nSigned-off-by: Hans Wippel \u003chwippel@linux.ibm.com\u003e\nSigned-off-by: Ursula Braun \u003cubraun@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1e2b1046b53f356e663fd145d0c57bea04a0343a", "tree": "7896873fdca8323b4dba75d0173edd42968b871e", "parents": [ "86de5921a3d5dd246df661e09bdd0a6131b39ae3", "1770f0fa978ed691a454fa9d3ba903d5a70630cc" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 15:51:16 2018 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 15:51:16 2018 -0800" }, "message": "Merge tag \u0027wireless-drivers-for-davem-2018-11-20\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers\n\nKalle Valo says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nwireless-drivers fixes for 4.20\n\nFirst set of fixes for 4.20, this time we have quite a few them but\nall very small.\n\nath9k\n\n* fix a locking regression found by a static checker\n\nwlcore\n\n* fix a crash which was a regression with wakeirq handling\n\nbrcm80211\n\n* yet another fix for 160 MHz channel handling\n\nmt76\n\n* fix a longstaning build problem when CONFIG_LEDS_CLASS is disabled\n\n* don\u0027t use uninitialised mutex\n\niwlwifi\n\n* do note that the iwlwifi merge tag (commit 4ec321c14693) seems to\n contain wrong list of changes so ignore that\n\n* fix ACPI data handling, a memory leak and other smaller fixes\n\nath10k\n\n* fix a crash during suspend which was a recent regression\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": "86de5921a3d5dd246df661e09bdd0a6131b39ae3", "tree": "efa8b73f929d9a2e136aaa0e0fa1605d6c12ed1e", "parents": [ "b5dd186d10ba59e6b5ba60e42b3b083df56df6f3" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Tue Nov 20 05:53:59 2018 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 15:49:52 2018 -0800" }, "message": "tcp: defer SACK compression after DupThresh\n\nJean-Louis reported a TCP regression and bisected to recent SACK\ncompression.\n\nAfter a loss episode (receiver not able to keep up and dropping\npackets because its backlog is full), linux TCP stack is sending\na single SACK (DUPACK).\n\nSender waits a full RTO timer before recovering losses.\n\nWhile RFC 6675 says in section 5, \"Algorithm Details\",\n\n (2) If DupAcks \u003c DupThresh but IsLost (HighACK + 1) returns true --\n indicating at least three segments have arrived above the current\n cumulative acknowledgment point, which is taken to indicate loss\n -- go to step (4).\n...\n (4) Invoke fast retransmit and enter loss recovery as follows:\n\nthere are old TCP stacks not implementing this strategy, and\nstill counting the dupacks before starting fast retransmit.\n\nWhile these stacks probably perform poorly when receivers implement\nLRO/GRO, we should be a little more gentle to them.\n\nThis patch makes sure we do not enable SACK compression unless\n3 dupacks have been sent since last rcv_nxt update.\n\nIdeally we should even rearm the timer to send one or two\nmore DUPACK if no more packets are coming, but that will\nbe work aiming for linux-4.21.\n\nMany thanks to Jean-Louis for bisecting the issue, providing\npacket captures and testing this patch.\n\nFixes: 5d9f4262b7ea (\"tcp: add SACK compression\")\nReported-by: Jean-Louis Dupond \u003cjean-louis@dupond.be\u003e\nTested-by: Jean-Louis Dupond \u003cjean-louis@dupond.be\u003e\nSigned-off-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": "b5dd186d10ba59e6b5ba60e42b3b083df56df6f3", "tree": "b6dd9d539cb37bdfd87b5ff873d179fae6c033ea", "parents": [ "f2cbd485282014132851bf37cb2ca624a456275d" ], "author": { "name": "Petr Machata", "email": "petrm@mellanox.com", "time": "Tue Nov 20 11:39:56 2018 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Nov 21 15:38:52 2018 -0800" }, "message": "net: skb_scrub_packet(): Scrub offload_fwd_mark\n\nWhen a packet is trapped and the corresponding SKB marked as\nalready-forwarded, it retains this marking even after it is forwarded\nacross veth links into another bridge. There, since it ingresses the\nbridge over veth, which doesn\u0027t have offload_fwd_mark, it triggers a\nwarning in nbp_switchdev_frame_mark().\n\nThen nbp_switchdev_allowed_egress() decides not to allow egress from\nthis bridge through another veth, because the SKB is already marked, and\nthe mark (of 0) of course matches. Thus the packet is incorrectly\nblocked.\n\nSolve by resetting offload_fwd_mark() in skb_scrub_packet(). That\nfunction is called from tunnels and also from veth, and thus catches the\ncases where traffic is forwarded between bridges and transformed in a\nway that invalidates the marking.\n\nFixes: 6bc506b4fb06 (\"bridge: switchdev: Add forward mark support for stacked devices\")\nFixes: abf4bb6b63d0 (\"skbuff: Add the offload_mr_fwd_mark field\")\nSigned-off-by: Petr Machata \u003cpetrm@mellanox.com\u003e\nSuggested-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\nAcked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "92b419289ceecdd1eae03114928913f298b84327", "tree": "3064c8f2c72e5438262a7e8968ae3b2071501117", "parents": [ "c8ce94b8fe5324aca4c10baeb53dea2d5775a170", "5d8f81ba1da55210123b9595e87b913c79579d02" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 21 11:28:20 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 21 11:28:20 2018 -0800" }, "message": "Merge tag \u0027riscv-for-linus-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux\n\nPull RISC-V fixes from Palmer Dabbelt:\n \"This week is a bit bigger than I expected. That\u0027s my fault, as I\n missed a few patches while I was at Plumbers last week. We have:\n\n - A fix to a quite embarassing issue where raw_copy_to_user() was\n implemented with asm_copy_from_user() (and vice versa).\n\n - Improvements to our makefile to allow flat binaries to be\n generated.\n\n - A build fix that predeclares \"struct module\" at the top of\n \u003casm/module.h\u003e, which triggers warnings later in that header.\n\n - The addition of our own \u003cuapi/asm/unistd\u003e header, which is\n necessary to align our stat ABI on 32-bit systems.\n\n - A fix to avoid printing a warning when the S or U bits are set in\n print_isa().\n\n I already have one patch in the queue for next week\"\n\n* tag \u0027riscv-for-linus-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux:\n RISC-V: recognize S/U mode bits in print_isa\n riscv: add asm/unistd.h UAPI header\n riscv: fix warning in arch/riscv/include/asm/module.h\n RISC-V: Build flat and compressed kernel images\n RISC-V: Fix raw_copy_{to,from}_user()\n" }, { "commit": "8c110d43c6bca4b24dd13272a9d4e0ba6f2ec957", "tree": "40d349bbd89a3f37587309c8f6f24ac122020a94", "parents": [ "494633fac7896afc2bce6f83fe7319946270540b" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Wed Nov 21 08:06:37 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Wed Nov 21 10:10:54 2018 -0800" }, "message": "iomap: readpages doesn\u0027t zero page tail beyond EOF\n\nWhen we read the EOF page of the file via readpages, we need\nto zero the region beyond EOF that we either do not read or\nshould not contain data so that mmap does not expose stale data to\nuser applications.\n\nHowever, iomap_adjust_read_range() fails to detect EOF correctly,\nand so fsx on 1k block size filesystems fails very quickly with\nmapreads exposing data beyond EOF. There are two problems here.\n\nFirstly, when calculating the end block of the EOF byte, we have\nto round the size by one to avoid a block aligned EOF from reporting\na block too large. i.e. a size of 1024 bytes is 1 block, which in\nindex terms is block 0. Therefore we have to calculate the end block\nfrom (isize - 1), not isize.\n\nThe second bug is determining if the current page spans EOF, and so\nwhether we need split it into two half, one for the IO, and the\nother for zeroing. Unfortunately, the code that checks whether\nwe should split the block doesn\u0027t actually check if we span EOF, it\njust checks if the read spans the /offset in the page/ that EOF\nsits on. So it splits every read into two if EOF is not page\naligned, regardless of whether we are reading the EOF block or not.\n\nHence we need to restrict the \"does the read span EOF\" check to\njust the page that spans EOF, not every page we read.\n\nThis patch results in correct EOF detection through readpages:\n\nxfs_vm_readpages: dev 259:0 ino 0x43 nr_pages 24\nxfs_iomap_found: dev 259:0 ino 0x43 size 0x66c00 offset 0x4f000 count 98304 type hole startoff 0x13c startblock 1368 blockcount 0x4\niomap_readpage_actor: orig pos 323584 pos 323584, length 4096, poff 0 plen 4096, isize 420864\nxfs_iomap_found: dev 259:0 ino 0x43 size 0x66c00 offset 0x50000 count 94208 type hole startoff 0x140 startblock 1497 blockcount 0x5c\niomap_readpage_actor: orig pos 327680 pos 327680, length 94208, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 331776 pos 331776, length 90112, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 335872 pos 335872, length 86016, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 339968 pos 339968, length 81920, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 344064 pos 344064, length 77824, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 348160 pos 348160, length 73728, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 352256 pos 352256, length 69632, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 356352 pos 356352, length 65536, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 360448 pos 360448, length 61440, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 364544 pos 364544, length 57344, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 368640 pos 368640, length 53248, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 372736 pos 372736, length 49152, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 376832 pos 376832, length 45056, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 380928 pos 380928, length 40960, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 385024 pos 385024, length 36864, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 389120 pos 389120, length 32768, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 393216 pos 393216, length 28672, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 397312 pos 397312, length 24576, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 401408 pos 401408, length 20480, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 405504 pos 405504, length 16384, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 409600 pos 409600, length 12288, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 413696 pos 413696, length 8192, poff 0 plen 4096, isize 420864\niomap_readpage_actor: orig pos 417792 pos 417792, length 4096, poff 0 plen 3072, isize 420864\niomap_readpage_actor: orig pos 420864 pos 420864, length 1024, poff 3072 plen 1024, isize 420864\n\nAs you can see, it now does full page reads until the last one which\nis split correctly at the block aligned EOF, reading 3072 bytes and\nzeroing the last 1024 bytes. The original version of the patch got\nthis right, but it got another case wrong.\n\nThe EOF detection crossing really needs to the the original length\nas plen, while it starts at the end of the block, will be shortened\nas up-to-date blocks are found on the page. This means \"orig_pos +\nplen\" no longer points to the end of the page, and so will not\ncorrectly detect EOF crossing. Hence we have to use the length\npassed in to detect this partial page case:\n\nxfs_filemap_fault: dev 259:1 ino 0x43 write_fault 0\nxfs_vm_readpage: dev 259:1 ino 0x43 nr_pages 1\nxfs_iomap_found: dev 259:1 ino 0x43 size 0x2cc00 offset 0x2c000 count 4096 type hole startoff 0xb0 startblock 282 blockcount 0x4\niomap_readpage_actor: orig pos 180224 pos 181248, length 4096, poff 1024 plen 2048, isize 183296\nxfs_iomap_found: dev 259:1 ino 0x43 size 0x2cc00 offset 0x2cc00 count 1024 type hole startoff 0xb3 startblock 285 blockcount 0x1\niomap_readpage_actor: orig pos 183296 pos 183296, length 1024, poff 3072 plen 1024, isize 183296\n\nHeere we see a trace where the first block on the EOF page is up to\ndate, hence poff \u003d 1024 bytes. The offset into the page of EOF is\n3072, so the range we want to read is 1024 - 3071, and the range we\nwant to zero is 3072 - 4095. You can see this is split correctly\nnow.\n\nThis fixes the stale data beyond EOF problem that fsx quickly\nuncovers on 1k block size filesystems.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "494633fac7896afc2bce6f83fe7319946270540b", "tree": "86c87a32c846defd14da23cface9d4f6a8ee677e", "parents": [ "4721a6010990971440b4ffefbdf014976b8eda2f" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:12 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Wed Nov 21 10:10:54 2018 -0800" }, "message": "vfs: vfs_dedupe_file_range() doesn\u0027t return EOPNOTSUPP\n\nIt returns EINVAL when the operation is not supported by the\nfilesystem. Fix it to return EOPNOTSUPP to be consistent with\nthe man page and clone_file_range().\n\nClean up the inconsistent error return handling while I\u0027m there.\n(I know, lipstick on a pig, but every little bit helps...)\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "4721a6010990971440b4ffefbdf014976b8eda2f", "tree": "a46df06954e1d079a2a592e5dccc7429fa2a9f8b", "parents": [ "b450672fb66b4a991a5b55ee24209ac7ae7690ce" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:11 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Wed Nov 21 10:10:53 2018 -0800" }, "message": "iomap: dio data corruption and spurious errors when pipes fill\n\nWhen doing direct IO to a pipe for do_splice_direct(), then pipe is\ntrivial to fill up and overflow as it can only hold 16 pages. At\nthis point bio_iov_iter_get_pages() then returns -EFAULT, and we\nabort the IO submission process. Unfortunately, iomap_dio_rw()\npropagates the error back up the stack.\n\nThe error is converted from the EFAULT to EAGAIN in\ngeneric_file_splice_read() to tell the splice layers that the pipe\nis full. do_splice_direct() completely fails to handle EAGAIN errors\n(it aborts on error) and returns EAGAIN to the caller.\n\ncopy_file_write() then completely fails to handle EAGAIN as well,\nand so returns EAGAIN to userspace, having failed to copy the data\nit was asked to.\n\nAvoid this whole steaming pile of fail by having iomap_dio_rw()\nsilently swallow EFAULT errors and so do short reads.\n\nTo make matters worse, iomap_dio_actor() has a stale data exposure\nbug bio_iov_iter_get_pages() fails - it does not zero the tail block\nthat it may have been left uncovered by partial IO. Fix the error\nhandling case to drop to the sub-block zeroing rather than\nimmmediately returning the -EFAULT error.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "b450672fb66b4a991a5b55ee24209ac7ae7690ce", "tree": "dcee5c6d6b1f32680c4b3a25741247dc1d6bcfe4", "parents": [ "0929d8580071c6a1cec1a7916a8f674c243ceee1" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:10 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Wed Nov 21 10:10:53 2018 -0800" }, "message": "iomap: sub-block dio needs to zeroout beyond EOF\n\nIf we are doing sub-block dio that extends EOF, we need to zero\nthe unused tail of the block to initialise the data in it it. If we\ndo not zero the tail of the block, then an immediate mmap read of\nthe EOF block will expose stale data beyond EOF to userspace. Found\nwith fsx running sub-block DIO sizes vs MAPREAD/MAPWRITE operations.\n\nFix this by detecting if the end of the DIO write is beyond EOF\nand zeroing the tail if necessary.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "0929d8580071c6a1cec1a7916a8f674c243ceee1", "tree": "665fe3dd2d8f33b072b2a1d62bd73517a7160b4e", "parents": [ "9230a0b65b47fe6856c4468ec0175c4987e5bede" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:10 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Wed Nov 21 10:10:53 2018 -0800" }, "message": "iomap: FUA is wrong for DIO O_DSYNC writes into unwritten extents\n\nWhen we write into an unwritten extent via direct IO, we dirty\nmetadata on IO completion to convert the unwritten extent to\nwritten. However, when we do the FUA optimisation checks, the inode\nmay be clean and so we issue a FUA write into the unwritten extent.\nThis means we then bypass the generic_write_sync() call after\nunwritten extent conversion has ben done and we don\u0027t force the\nmodified metadata to stable storage.\n\nThis violates O_DSYNC semantics. The window of exposure is a single\nIO, as the next DIO write will see the inode has dirty metadata and\nhence will not use the FUA optimisation. Calling\ngeneric_write_sync() after completion of the second IO will also\nsync the first write and it\u0027s metadata.\n\nFix this by avoiding the FUA optimisation when writing to unwritten\nextents.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "9230a0b65b47fe6856c4468ec0175c4987e5bede", "tree": "5b3a38652031d6053361af4dedbc62b3ad69e7ef", "parents": [ "2c307174ab77e34645e75e12827646e044d273c3" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 22:50:08 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Wed Nov 21 10:10:53 2018 -0800" }, "message": "xfs: delalloc -\u003e unwritten COW fork allocation can go wrong\n\nLong saga. There have been days spent following this through dead end\nafter dead end in multi-GB event traces. This morning, after writing\na trace-cmd wrapper that enabled me to be more selective about XFS\ntrace points, I discovered that I could get just enough essential\ntracepoints enabled that there was a 50:50 chance the fsx config\nwould fail at ~115k ops. If it didn\u0027t fail at op 115547, I stopped\nfsx at op 115548 anyway.\n\nThat gave me two traces - one where the problem manifested, and one\nwhere it didn\u0027t. After refining the traces to have the necessary\ninformation, I found that in the failing case there was a real\nextent in the COW fork compared to an unwritten extent in the\nworking case.\n\nWalking back through the two traces to the point where the CWO fork\nextents actually diverged, I found that the bad case had an extra\nunwritten extent in it. This is likely because the bug it led me to\nhad triggered multiple times in those 115k ops, leaving stray\nCOW extents around. What I saw was a COW delalloc conversion to an\nunwritten extent (as they should always be through\nxfs_iomap_write_allocate()) resulted in a /written extent/:\n\nxfs_writepage: dev 259:0 ino 0x83 pgoff 0x17000 size 0x79a00 offset 0 length 0\nxfs_iext_remove: dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/2 offset 32 block 152 count 20 flag 1 caller xfs_bmap_add_extent_delay_real\nxfs_bmap_pre_update: dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/1 offset 1 block 4503599627239429 count 31 flag 0 caller xfs_bmap_add_extent_delay_real\nxfs_bmap_post_update: dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/1 offset 1 block 121 count 51 flag 0 caller xfs_bmap_add_ex\n\nBasically, Cow fork before:\n\n\t0 1 32 52\n\t+H+DDDDDDDDDDDD+UUUUUUUUUUU+\n\t PREV\t\tRIGHT\n\nCOW delalloc conversion allocates:\n\n\t 1\t 32\n\t +uuuuuuuuuuuu+\n\t NEW\n\nAnd the result according to the xfs_bmap_post_update trace was:\n\n\t0 1 32 52\n\t+H+wwwwwwwwwwwwwwwwwwwwwwww+\n\t PREV\n\nWhich is clearly wrong - it should be a merged unwritten extent,\nnot an unwritten extent.\n\nThat lead me to look at the LEFT_FILLING|RIGHT_FILLING|RIGHT_CONTIG\ncase in xfs_bmap_add_extent_delay_real(), and sure enough, there\u0027s\nthe bug.\n\nIt takes the old delalloc extent (PREV) and adds the length of the\nRIGHT extent to it, takes the start block from NEW, removes the\nRIGHT extent and then updates PREV with the new extent.\n\nWhat it fails to do is update PREV.br_state. For delalloc, this is\nalways XFS_EXT_NORM, while in this case we are converting the\ndelayed allocation to unwritten, so it needs to be updated to\nXFS_EXT_UNWRITTEN. This LF|RF|RC case does not do this, and so\nthe resultant extent is always written.\n\nAnd that\u0027s the bug I\u0027ve been chasing for a week - a bmap btree bug,\nnot a reflink/dedupe/copy_file_range bug, but a BMBT bug introduced\nwith the recent in core extent tree scalability enhancements.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "2c307174ab77e34645e75e12827646e044d273c3", "tree": "6daad4c4a35de98f1c27b54254880d9c49707049", "parents": [ "7f9f71be84bcab368e58020a42f6d0dd97adf0ce" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:10 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Wed Nov 21 10:10:53 2018 -0800" }, "message": "xfs: flush removing page cache in xfs_reflink_remap_prep\n\nOn a sub-page block size filesystem, fsx is failing with a data\ncorruption after a series of operations involving copying a file\nwith the destination offset beyond EOF of the destination of the file:\n\n8093(157 mod 256): TRUNCATE DOWN from 0x7a120 to 0x50000 ******WWWW\n8094(158 mod 256): INSERT 0x25000 thru 0x25fff (0x1000 bytes)\n8095(159 mod 256): COPY 0x18000 thru 0x1afff (0x3000 bytes) to 0x2f400\n8096(160 mod 256): WRITE 0x5da00 thru 0x651ff (0x7800 bytes) HOLE\n8097(161 mod 256): COPY 0x2000 thru 0x5fff (0x4000 bytes) to 0x6fc00\n\nThe second copy here is beyond EOF, and it is to sub-page (4k) but\nblock aligned (1k) offset. The clone runs the EOF zeroing, landing\nin a pre-existing post-eof delalloc extent. This zeroes the post-eof\nextents in the page cache just fine, dirtying the pages correctly.\n\nThe problem is that xfs_reflink_remap_prep() now truncates the page\ncache over the range that it is copying it to, and rounds that down\nto cover the entire start page. This removes the dirty page over the\ndelalloc extent from the page cache without having written it back.\nHence later, when the page cache is flushed, the page at offset\n0x6f000 has not been written back and hence exposes stale data,\nwhich fsx trips over less than 10 operations later.\n\nFix this by changing xfs_reflink_remap_prep() to use\nxfs_flush_unmap_range().\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "cb216b84d6ea24fa10f1e7aac35de77246841041", "tree": "4c1d102f7214ba97b3a43a1af8ba9495e23f3d76", "parents": [ "b34087157dd76e8d96e5e52808134a791ac61e57" ], "author": { "name": "Robin Murphy", "email": "robin.murphy@arm.com", "time": "Wed Nov 21 16:00:51 2018 +0000" }, "committer": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Nov 21 18:47:58 2018 +0100" }, "message": "swiotlb: Skip cache maintenance on map error\n\nIf swiotlb_bounce_page() failed, calling arch_sync_dma_for_device() may\nlead to such delights as performing cache maintenance on whatever\naddress phys_to_virt(SWIOTLB_MAP_ERROR) looks like, which is typically\noutside the kernel memory map and goes about as well as expected.\n\nDon\u0027t do that.\n\nFixes: a4a4330db46a (\"swiotlb: add support for non-coherent DMA\")\nTested-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\nSigned-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n" }, { "commit": "b34087157dd76e8d96e5e52808134a791ac61e57", "tree": "94aba2a4c0adae46be45c150ff5a57f15805d882", "parents": [ "c8ce94b8fe5324aca4c10baeb53dea2d5775a170" ], "author": { "name": "Robin Murphy", "email": "robin.murphy@arm.com", "time": "Wed Nov 21 16:00:50 2018 +0000" }, "committer": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Nov 21 18:47:52 2018 +0100" }, "message": "dma-direct: Make DIRECT_MAPPING_ERROR viable for SWIOTLB\n\nWith the overflow buffer removed, we no longer have a unique address\nwhich is guaranteed not to be a valid DMA target to use as an error\ntoken. The DIRECT_MAPPING_ERROR value of 0 tries to at least represent\nan unlikely DMA target, but unfortunately there are already SWIOTLB\nusers with DMA-able memory at physical address 0 which now gets falsely\ntreated as a mapping failure and leads to all manner of misbehaviour.\n\nThe best we can do to mitigate that is flip DIRECT_MAPPING_ERROR to the\nother commonly-used error value of all-bits-set, since the last single\nbyte of memory is by far the least-likely-valid DMA target.\n\nFixes: dff8d6c1ed58 (\"swiotlb: remove the overflow buffer\")\nReported-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\nTested-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\nAcked-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nSigned-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n" }, { "commit": "14b04063cc994effc86f976625bf8f806d8d44cb", "tree": "b250b7a408e8ffb1a51346fca54301d76aaebf22", "parents": [ "8dc765d438f1e42b3e8227b3b09fad7d73f4ec9a", "4cff280a5fccf6513ed9e895bb3a4e7ad8b0cedc" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Nov 21 05:56:28 2018 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Nov 21 05:56:28 2018 -0700" }, "message": "Merge branch \u0027nvme-4.20\u0027 of git://git.infradead.org/nvme into for-linus\n\nPull NVMe fix from Christoph.\n\n* \u0027nvme-4.20\u0027 of git://git.infradead.org/nvme:\n nvme-fc: resolve io failures during connect\n" }, { "commit": "0db699f747cc24835af77a34852882a33b3e1fc6", "tree": "3e02733a7e5987ab5c771ca98c4180482d0c449e", "parents": [ "9ff01193a20d391e8dbce4403dd5ef87c7eaaca6", "dbc4ca339c8dbdd8652ce57c16bf5ef45ee4307e" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Nov 21 13:33:06 2018 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Nov 21 13:33:06 2018 +0100" }, "message": "Merge tag \u0027linux-cpupower-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux\n\nPull cpupower utility updates for 4.20-rc4 from Shuah Khan:\n\n\"This cpupower update for Linux 4.20-rc4 consists of compile fixes to allow\n use of outside build flags and override of CFLAGS from Jiri Olsa, and fix\n to compilation with STATIC\u003dtrue from Konstantin Khlebnikov.\"\n\n* tag \u0027linux-cpupower-4.20-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux:\n tools cpupower: Override CFLAGS assignments\n tools cpupower debug: Allow to use outside build flags\n tools/power/cpupower: fix compilation with STATIC\u003dtrue\n" }, { "commit": "f559156c399cfb11d53a128d210118fbea36816e", "tree": "c7b64ccbe688eb8e656bb692401c74bf36f665cc", "parents": [ "c773058dde9a4f919a8069f3828d9f4adb1fce1e" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Tue Nov 20 15:54:50 2018 +0200" }, "committer": { "name": "Joonas Lahtinen", "email": "joonas.lahtinen@linux.intel.com", "time": "Wed Nov 21 14:30:58 2018 +0200" }, "message": "drm/i915: Add rotation readout for plane initial config\n\nIf we need to force a full plane update before userspace/fbdev\nhave given us a proper plane state we should try to maintain the\ncurrent plane state as much as possible (apart from the parts\nof the state we\u0027re trying to fix up with the plane update).\nTo that end add basic readout for the plane rotation and\nmaintain it during the initial fb takeover.\n\nCc: Hans de Goede \u003chdegoede@redhat.com\u003e\nFixes: 516a49cc1946 (\"drm/i915: Fix assert_plane() warning on bootup with external display\")\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20181120135450.3634-2-ville.syrjala@linux.intel.com\nTested-by: Hans de Goede \u003chdegoede@redhat.com\u003e\nReviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nReviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n(cherry picked from commit f43348a3db89305bb1935da9fe4499fdcdde9796)\nSigned-off-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n" }, { "commit": "c773058dde9a4f919a8069f3828d9f4adb1fce1e", "tree": "23a33a9058e157b9fc13c7745a107c04b6e61c43", "parents": [ "f8577fb3c273bcae821a5254c8fbcf82016d9a8d" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Tue Nov 20 15:54:49 2018 +0200" }, "committer": { "name": "Joonas Lahtinen", "email": "joonas.lahtinen@linux.intel.com", "time": "Wed Nov 21 14:30:54 2018 +0200" }, "message": "drm/i915: Force a LUT update in intel_initial_commit()\n\nIf we force a plane update to fix up our half populated plane state\nwe\u0027ll also force on the pipe gamma for the plane (since we always\nenable pipe gamma currently). If the BIOS hasn\u0027t programmed a sensible\nLUT into the hardware this will cause the image to become corrupted.\nTypical symptoms are a purple/yellow/etc. flash when the driver loads.\n\nTo avoid this let\u0027s program something sensible into the LUT when\nwe do the plane update. In the future I plan to add proper plane\ngamma enable readout so this is just a temporary measure.\n\nCc: Hans de Goede \u003chdegoede@redhat.com\u003e\nFixes: 516a49cc1946 (\"drm/i915: Fix assert_plane() warning on bootup with external display\")\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20181120135450.3634-1-ville.syrjala@linux.intel.com\nTested-by: Hans de Goede \u003chdegoede@redhat.com\u003e\nReviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n(cherry picked from commit fa6af5145b4e87a30a530be0d80734a9dd40da77)\nSigned-off-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n" }, { "commit": "2bbb5fa37475d7aa5fa62f34db1623f3da2dfdfa", "tree": "149ab13572d511ae694d1004570a54f1240c9128", "parents": [ "9ff01193a20d391e8dbce4403dd5ef87c7eaaca6" ], "author": { "name": "Hans de Goede", "email": "hdegoede@redhat.com", "time": "Mon Nov 19 19:06:01 2018 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Nov 21 13:30:13 2018 +0100" }, "message": "ACPI / platform: Add SMB0001 HID to forbidden_id_list\n\nMany HP AMD based laptops contain an SMB0001 device like this:\n\nDevice (SMBD)\n{\n Name (_HID, \"SMB0001\") // _HID: Hardware ID\n Name (_CRS, ResourceTemplate () // _CRS: Current Resource Settings\n {\n IO (Decode16,\n 0x0B20, // Range Minimum\n 0x0B20, // Range Maximum\n 0x20, // Alignment\n 0x20, // Length\n )\n IRQ (Level, ActiveLow, Shared, )\n {7}\n })\n}\n\nThe legacy style IRQ resource here causes acpi_dev_get_irqresource() to\nbe called with legacy\u003dtrue and this message to show in dmesg:\nACPI: IRQ 7 override to edge, high\n\nThis causes issues when later on the AMD0030 GPIO device gets enumerated:\n\nDevice (GPIO)\n{\n Name (_HID, \"AMDI0030\") // _HID: Hardware ID\n Name (_CID, \"AMDI0030\") // _CID: Compatible ID\n Name (_UID, Zero) // _UID: Unique ID\n Method (_CRS, 0, NotSerialized) // _CRS: Current Resource Settings\n {\n\tName (RBUF, ResourceTemplate ()\n\t{\n\t Interrupt (ResourceConsumer, Level, ActiveLow, Shared, ,, )\n\t {\n\t\t0x00000007,\n\t }\n\t Memory32Fixed (ReadWrite,\n\t\t0xFED81500, // Address Base\n\t\t0x00000400, // Address Length\n\t\t)\n\t})\n\tReturn (RBUF) /* \\_SB_.GPIO._CRS.RBUF */\n }\n}\n\nNow acpi_dev_get_irqresource() gets called with legacy\u003dfalse, but because\nof the earlier override of the trigger-type acpi_register_gsi() returns\n-EBUSY (because we try to register the same interrupt with a different\ntrigger-type) and we end up setting IORESOURCE_DISABLED in the flags.\n\nThe setting of IORESOURCE_DISABLED causes platform_get_irq() to call\nacpi_irq_get() which is not implemented on x86 and returns -EINVAL.\nresulting in the following in dmesg:\n\namd_gpio AMDI0030:00: Failed to get gpio IRQ: -22\namd_gpio: probe of AMDI0030:00 failed with error -22\n\nThe SMB0001 is a \"virtual\" device in the sense that the only way the OS\ninteracts with it is through calling a couple of methods to do SMBus\ntransfers. As such it is weird that it has IO and IRQ resources at all,\nbecause the driver for it is not expected to ever access the hardware\ndirectly.\n\nThe Linux driver for the SMB0001 device directly binds to the acpi_device\nthrough the acpi_bus, so we do not need to instantiate a platform_device\nfor this ACPI device. This commit adds the SMB0001 HID to the\nforbidden_id_list, avoiding the instantiating of a platform_device for it.\nNot instantiating a platform_device means we will no longer call\nacpi_dev_get_irqresource() for the legacy IRQ resource fixing the probe of\nthe AMDI0030 device failing.\n\nBugLink: https://bugzilla.redhat.com/show_bug.cgi?id\u003d1644013\nBugLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d198715\nBugLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d199523\nReported-by: Lukas Kahnert \u003copenproggerfreak@gmail.com\u003e\nTested-by: Marc \u003csuaefar@googlemail.com\u003e\nCc: All applicable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n" }, { "commit": "8fd3b90300bec541806dac271de2fd44e2e4e2d2", "tree": "d71300fce2291a4ba03b625d1c56e106f4751726", "parents": [ "5c074eeabbd332b11559f7fc1e89d456f94801fb" ], "author": { "name": "Paul Kocialkowski", "email": "paul.kocialkowski@bootlin.com", "time": "Thu Nov 15 17:32:48 2018 +0100" }, "committer": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Wed Nov 21 10:38:19 2018 +0100" }, "message": "drm/fb-helper: Blacklist writeback when adding connectors to fbdev\n\nWriteback connectors do not produce any on-screen output and require\nspecial care for use. Such connectors are hidden from enumeration in\nDRM resources by default, but they are still picked-up by fbdev.\nThis makes rather little sense since fbdev is not really adapted for\ndealing with writeback.\n\nMoreover, this is also a source of issues when userspace disables the\nCRTC (and associated plane) without detaching the CRTC from the\nconnector (which is hidden by default). In this case, the connector is\nstill using the CRTC, leading to am \"enabled/connectors mismatch\" and\neventually the failure of the associated atomic commit. This situation\nhappens with VC4 testing under IGT GPU Tools.\n\nFilter out writeback connectors in the fbdev helper to solve this.\n\nSigned-off-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\nReviewed-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nReviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nTested-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nFixes: 935774cd71fe (\"drm: Add writeback connector type\")\nCc: \u003cstable@vger.kernel.org\u003e # v4.19+\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20181115163248.21168-1-paul.kocialkowski@bootlin.com\n" }, { "commit": "f8577fb3c273bcae821a5254c8fbcf82016d9a8d", "tree": "39a142c18da0af09bccce9c11b3ccbde5a3a1b99", "parents": [ "8830f26bcd3cf1ff07d9078cd310a534c03b6a10" ], "author": { "name": "Chris Wilson", "email": "chris@chris-wilson.co.uk", "time": "Mon Nov 19 15:41:53 2018 +0000" }, "committer": { "name": "Joonas Lahtinen", "email": "joonas.lahtinen@linux.intel.com", "time": "Wed Nov 21 09:32:08 2018 +0200" }, "message": "drm/i915: Write GPU relocs harder with gen3\n\nUnder moderate amounts of GPU stress, we can observe on Bearlake and\nPineview (later gen3 models) that we execute the following batch buffer\nbefore the write into the batch is coherent. Adding extra (tested with\nupto 32x) MI_FLUSH to either the invalidation, flush or both phases does\nnot solve the incoherency issue with the relocations, but emitting the\nMI_STORE_DWORD_IMM twice does. So be it.\n\nFixes: 7dd4f6729f92 (\"drm/i915: Async GPU relocation processing\")\nTestcase: igt/gem_tiled_fence_blits # blb/pnv\nSigned-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\nReviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20181119154153.15327-1-chris@chris-wilson.co.uk\n(cherry picked from commit 7fa28e146994da1e8a4124623d7da97b798ea520)\nSigned-off-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n" }, { "commit": "f2cbd485282014132851bf37cb2ca624a456275d", "tree": "60b84978c054decb6e70e7dd99e294220d347c64", "parents": [ "b1d982330134bbba371821bc850da6fa4bfa3271" ], "author": { "name": "Davide Caratti", "email": "dcaratti@redhat.com", "time": "Tue Nov 20 22:18:44 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 20 14:59:58 2018 -0800" }, "message": "net/sched: act_police: fix race condition on state variables\n\nafter \u0027police\u0027 configuration parameters were converted to use RCU instead\nof spinlock, the state variables used to compute the traffic rate (namely\n\u0027tcfp_toks\u0027, \u0027tcfp_ptoks\u0027 and \u0027tcfp_t_c\u0027) are erroneously read/updated in\nthe traffic path without any protection.\n\nUse a dedicated spinlock to avoid race conditions on these variables, and\nensure proper cache-line alignment. In this way, \u0027police\u0027 is still faster\nthan what we observed when \u0027tcf_lock\u0027 was used in the traffic path _ i.e.\nreverting commit 2d550dbad83c (\"net/sched: act_police: don\u0027t use spinlock\nin the data path\"). Moreover, we preserve the throughput improvement that\nwas obtained after \u0027police\u0027 started using per-cpu counters, when \u0027avrate\u0027\nis used instead of \u0027rate\u0027.\n\nChanges since v1 (thanks to Eric Dumazet):\n- call ktime_get_ns() before acquiring the lock in the traffic path\n- use a dedicated spinlock instead of tcf_lock\n- improve cache-line usage\n\nFixes: 2d550dbad83c (\"net/sched: act_police: don\u0027t use spinlock in the data path\")\nReported-and-suggested-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: Davide Caratti \u003cdcaratti@redhat.com\u003e\nReviewed-by: Eric Dumazet \u003cedumazet@google.com\u003e\n" }, { "commit": "c8ce94b8fe5324aca4c10baeb53dea2d5775a170", "tree": "3e8c9eea0080a1a48f0d41782693f2dfcdedf275", "parents": [ "06e68fed32826b44aa9ffcf282c14d3c58918a70", "1229ace4a4a2e2c982a32fb075dc1bf95423924f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 20 14:31:00 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 20 14:31:00 2018 -0800" }, "message": "Merge tag \u0027mips_fixes_4.20_3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux\n\nPull MIPS fixes from Paul Burton:\n \"A few MIPS fixes for 4.20:\n\n - Re-enable the Cavium Octeon USB driver in its defconfig after it\n was accidentally removed back in 4.14.\n\n - Have early memblock allocations be performed bottom-up to more\n closely match the behaviour we used to have with bootmem, which\n seems a safer choice since we\u0027ve seen fallout from the change made\n in the 4.20 merge window.\n\n - Simplify max_low_pfn calculation in the NUMA code for the Loongson3\n and SGI IP27 platforms to both clean up the code \u0026 ensure\n max_low_pfn has been set appropriately before it is used\"\n\n* tag \u0027mips_fixes_4.20_3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:\n MIPS: Loongson3,SGI-IP27: Simplify max_low_pfn calculation\n MIPS: Let early memblock_alloc*() allocate memories bottom-up\n MIPS: OCTEON: cavium_octeon_defconfig: re-enable OCTEON USB driver\n" }, { "commit": "b1d982330134bbba371821bc850da6fa4bfa3271", "tree": "90409746061cd3e3f7e08f1b4bb4806314ea81e1", "parents": [ "cadf9df27e7cf40e390e060a1c71bb86ecde798b" ], "author": { "name": "Heiner Kallweit", "email": "hkallweit1@gmail.com", "time": "Tue Nov 20 21:22:50 2018 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 20 14:22:45 2018 -0800" }, "message": "MAINTAINERS: add myself as co-maintainer for r8169\n\nMeanwhile I know the driver quite well and I refactored bigger parts\nof it. As a result people contact me already with r8169 questions.\nTherefore I\u0027d volunteer to become co-maintainer of the driver also\nofficially.\n\nSigned-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a5d0f4565996e5595a10cb57b3d1e3d74379c502", "tree": "3ea81cc422815c3c43caf6e19d42c5ba35de30b0", "parents": [ "a4233cc944d1b7125d906f1fa276bda3df48df0c" ], "author": { "name": "Kenneth Feng", "email": "kenneth.feng@amd.com", "time": "Mon Nov 19 14:49:16 2018 +0800" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Nov 20 14:40:15 2018 -0500" }, "message": "drm/amdgpu: Enable HDP memory light sleep\n\nDue to the register name and setting change of HDP\nmemory light sleep on Vega20,change accordingly in\nthe driver.\n\nSigned-off-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\nReviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "7f9f71be84bcab368e58020a42f6d0dd97adf0ce", "tree": "424a0ea984721b50be1e95d00e019bf88791b4ad", "parents": [ "c08768977b9a65cab9bcfd1ba30ffb686b2b7c69" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:09 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Tue Nov 20 10:36:19 2018 -0800" }, "message": "xfs: extent shifting doesn\u0027t fully invalidate page cache\n\nThe extent shifting code uses a flush and invalidate mechainsm prior\nto shifting extents around. This is similar to what\nxfs_free_file_space() does, but it doesn\u0027t take into account things\nlike page cache vs block size differences, and it will fail if there\nis a page that it currently busy.\n\nxfs_flush_unmap_range() handles all of these cases, so just convert\nxfs_prepare_shift() to us that mechanism rather than having it\u0027s own\nspecial sauce.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "c08768977b9a65cab9bcfd1ba30ffb686b2b7c69", "tree": "311d571609f12baabb0bc7e85af18642194a84b4", "parents": [ "d43aaf1685aa471f0593685c9f54d53e3af3cf3f" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:08 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Tue Nov 20 10:36:11 2018 -0800" }, "message": "xfs: finobt AG reserves don\u0027t consider last AG can be a runt\n\nThe last AG may be very small comapred to all other AGs, and hence\nAG reservations based on the superblock AG size may actually consume\nmore space than the AG actually has. This results on assert failures\nlike:\n\nXFS: Assertion failed: xfs_perag_resv(pag, XFS_AG_RESV_METADATA)-\u003ear_reserved + xfs_perag_resv(pag, XFS_AG_RESV_RMAPBT)-\u003ear_reserved \u003c\u003d pag-\u003epagf_freeblks + pag-\u003epagf_flcount, file: fs/xfs/libxfs/xfs_ag_resv.c, line: 319\n[ 48.932891] xfs_ag_resv_init+0x1bd/0x1d0\n[ 48.933853] xfs_fs_reserve_ag_blocks+0x37/0xb0\n[ 48.934939] xfs_mountfs+0x5b3/0x920\n[ 48.935804] xfs_fs_fill_super+0x462/0x640\n[ 48.936784] ? xfs_test_remount_options+0x60/0x60\n[ 48.937908] mount_bdev+0x178/0x1b0\n[ 48.938751] mount_fs+0x36/0x170\n[ 48.939533] vfs_kern_mount.part.43+0x54/0x130\n[ 48.940596] do_mount+0x20e/0xcb0\n[ 48.941396] ? memdup_user+0x3e/0x70\n[ 48.942249] ksys_mount+0xba/0xd0\n[ 48.943046] __x64_sys_mount+0x21/0x30\n[ 48.943953] do_syscall_64+0x54/0x170\n[ 48.944835] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n\nHence we need to ensure the finobt per-ag space reservations take\ninto account the size of the last AG rather than treat it like all\nthe other full size AGs.\n\nNote that both refcountbt and rmapbt already take the size of the AG\ninto account via reading the AGF length directly.\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "d43aaf1685aa471f0593685c9f54d53e3af3cf3f", "tree": "bba393dac291862f520d931d9968fb8c4a030a9d", "parents": [ "d61fa8cbf3da85ffca6620f261354941c126ee23" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:08 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Tue Nov 20 10:36:01 2018 -0800" }, "message": "xfs: fix transient reference count error in xfs_buf_resubmit_failed_buffers\n\nWhen retrying a failed inode or dquot buffer,\nxfs_buf_resubmit_failed_buffers() clears all the failed flags from\nthe inde/dquot log items. In doing so, it also drops all the\nreference counts on the buffer that the failed log items hold. This\nmeans it can drop all the active references on the buffer and hence\nfree the buffer before it queues it for write again.\n\nPutting the buffer on the delwri queue takes a reference to the\nbuffer (so that it hangs around until it has been written and\ncompleted), but this goes bang if the buffer has already been freed.\n\nHence we need to add the buffer to the delwri queue before we remove\nthe failed flags from the log items attached to the buffer to ensure\nit always remains referenced during the resubmit process.\n\nReported-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "d61fa8cbf3da85ffca6620f261354941c126ee23", "tree": "14bb18b2ee7e26bc5b85c6b07561a8484da00500", "parents": [ "da034bcc6aaaf2a6ba6c5b5e63565c5ef4816a0e" ], "author": { "name": "Dave Chinner", "email": "dchinner@redhat.com", "time": "Mon Nov 19 13:31:07 2018 -0800" }, "committer": { "name": "Darrick J. Wong", "email": "darrick.wong@oracle.com", "time": "Tue Nov 20 10:35:05 2018 -0800" }, "message": "xfs: uncached buffer tracing needs to print bno\n\nUseless:\n\nxfs_buf_get_uncached: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_unlock: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_submit: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_hold: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_iowait: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_iodone: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_iowait_done: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_rele: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\n\nUseful:\n\n\nxfs_buf_get_uncached: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_unlock: dev 253:32 bno 0xffffffffffffffff nblks 0x1 ...\nxfs_buf_submit: dev 253:32 bno 0x200b5 nblks 0x1 ...\nxfs_buf_hold: dev 253:32 bno 0x200b5 nblks 0x1 ...\nxfs_buf_iowait: dev 253:32 bno 0x200b5 nblks 0x1 ...\nxfs_buf_iodone: dev 253:32 bno 0x200b5 nblks 0x1 ...\nxfs_buf_iowait_done: dev 253:32 bno 0x200b5 nblks 0x1 ...\nxfs_buf_rele: dev 253:32 bno 0x200b5 nblks 0x1 ...\n\nSigned-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n" }, { "commit": "cadf9df27e7cf40e390e060a1c71bb86ecde798b", "tree": "00e376aa0bc788a78b3c12242d914c0aeeccea8f", "parents": [ "59663e42199c93d1d7314d1446f6782fc4b1eb81" ], "author": { "name": "Stephen Mallon", "email": "stephen.mallon@sydney.edu.au", "time": "Tue Nov 20 19:15:02 2018 +1100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 20 10:32:11 2018 -0800" }, "message": "tcp: Fix SOF_TIMESTAMPING_RX_HARDWARE to use the latest timestamp during TCP coalescing\n\nDuring tcp coalescing ensure that the skb hardware timestamp refers to the\nhighest sequence number data.\nPreviously only the software timestamp was updated during coalescing.\n\nSigned-off-by: Stephen Mallon \u003cstephen.mallon@sydney.edu.au\u003e\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "59663e42199c93d1d7314d1446f6782fc4b1eb81", "tree": "d0286dafbd2bc3e30af491dd7adc051f23351ad2", "parents": [ "1359f25106cad6535812f900625150379c51df82" ], "author": { "name": "Siva Reddy Kallam", "email": "siva.kallam@broadcom.com", "time": "Tue Nov 20 10:04:04 2018 +0530" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Nov 20 10:18:05 2018 -0800" }, "message": "tg3: Add PHY reset for 5717/5719/5720 in change ring and flow control paths\n\nThis patch has the fix to avoid PHY lockup with 5717/5719/5720 in change\nring and flow control paths. This patch solves the RX hang while doing\ncontinuous ring or flow control parameters with heavy traffic from peer.\n\nSigned-off-by: Siva Reddy Kallam \u003csiva.kallam@broadcom.com\u003e\nAcked-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "544b03da39e2d7b4961d3163976ed4bfb1fac509", "tree": "fb73648550f9605147355cd31fa103afb6d334fe", "parents": [ "cb5d21946d2a2f4687c482ab4604af1d29dac35a" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Mon Nov 19 11:07:18 2018 +0000" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Nov 20 18:02:45 2018 +0100" }, "message": "Documentation/security-bugs: Postpone fix publication in exceptional cases\n\nAt the request of the reporter, the Linux kernel security team offers to\npostpone the publishing of a fix for up to 5 business days from the date\nof a report.\n\nWhile it is generally undesirable to keep a fix private after it has\nbeen developed, this short window is intended to allow distributions to\npackage the fix into their kernel builds and permits early inclusion of\nthe security team in the case of a co-ordinated disclosure with other\nparties. Unfortunately, discussions with major Linux distributions and\ncloud providers has revealed that 5 business days is not sufficient to\nachieve either of these two goals.\n\nAs an example, cloud providers need to roll out KVM security fixes to a\nglobal fleet of hosts with sufficient early ramp-up and monitoring. An\nend-to-end timeline of less than two weeks dramatically cuts into the\namount of early validation and increases the chance of guest-visible\nregressions.\n\nThe consequence of this timeline mismatch is that security issues are\ncommonly fixed without the involvement of the Linux kernel security team\nand are instead analysed and addressed by an ad-hoc group of developers\nacross companies contributing to Linux. In some cases, mainline (and\ntherefore the official stable kernels) can be left to languish for\nextended periods of time. This undermines the Linux kernel security\nprocess and puts upstream developers in a difficult position should they\nfind themselves involved with an undisclosed security problem that they\nare unable to report due to restrictions from their employer.\n\nTo accommodate the needs of these users of the Linux kernel and\nencourage them to engage with the Linux security team when security\nissues are first uncovered, extend the maximum period for which fixes\nmay be delayed to 7 calendar days, or 14 calendar days in exceptional\ncases, where the logistics of QA and large scale rollouts specifically\nneed to be accommodated. This brings parity with the linux-distros@\nmaximum embargo period of 14 calendar days.\n\nCc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nCc: David Woodhouse \u003cdwmw@amazon.co.uk\u003e\nCc: Amit Shah \u003caams@amazon.com\u003e\nCc: Laura Abbott \u003clabbott@redhat.com\u003e\nAcked-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCo-developed-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCo-developed-by: David Woodhouse \u003cdwmw@amazon.co.uk\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: David Woodhouse \u003cdwmw@amazon.co.uk\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nReviewed-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\nAcked-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "cb5d21946d2a2f4687c482ab4604af1d29dac35a", "tree": "b5fd5a321033898593912d96efee709dabf29869", "parents": [ "5082a7df5251b361be8f8edc1f0a148d07999820" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Thu Nov 15 15:03:24 2018 -0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue Nov 20 18:02:45 2018 +0100" }, "message": "MAINTAINERS: Add Sasha as a stable branch maintainer\n\nSasha has somehow been convinced into helping me with the stable kernel\nmaintenance. Codify this slip in good judgement before he realizes what\nhe really signed up for :)\n\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nAcked-by: Sasha Levin \u003csashal@kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "06e68fed32826b44aa9ffcf282c14d3c58918a70", "tree": "a15c80458660be89ce02d322be9edbb69136274f", "parents": [ "f2ce1065e767fc7da106a5f5381d1e8f842dc6f4", "4e26f692e2e2aa4d7d6ddb3c4d3dec17f45d6495" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 20 07:37:15 2018 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 20 07:37:15 2018 -0800" }, "message": "Merge tag \u0027media/v4.20-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media\n\nPull media fixes from Mauro Carvalho Chehab:\n\n - add a missing include at v4l2-controls uAPI header\n\n - minor kAPI update for the request API\n\n - some fixes at CEC core\n\n - use a lower minimum height for the virtual codec driver\n\n - cleanup a gcc warning due to the lack of a fall though markup\n\n - tc358743: Remove unnecessary self assignment\n\n - fix the V4L event subscription logic\n\n - docs: Document metadata format in struct v4l2_format\n\n - omap3isp and ipu3-cio2: fix unbinding logic\n\n* tag \u0027media/v4.20-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:\n media: ipu3-cio2: Use cio2_queues_exit\n media: ipu3-cio2: Unregister device nodes first, then release resources\n media: omap3isp: Unregister media device as first\n media: docs: Document metadata format in struct v4l2_format\n media: v4l: event: Add subscription to list before calling \"add\" operation\n media: dm365_ipipeif: better annotate a fall though\n media: Rename vb2_m2m_request_queue -\u003e v4l2_m2m_request_queue\n media: cec: increase debug level for \u0027queue full\u0027\n media: cec: check for non-OK/NACK conditions while claiming a LA\n media: vicodec: lower minimum height to 360\n media: tc358743: Remove unnecessary self assignment\n media: v4l: fix uapi mpeg slice params definition\n v4l2-controls: add a missing include\n" } ], "next": "e8828ec1c003727fc001eab06aa19bd2ca9b677e" }