)]}' { "log": [ { "commit": "b0c3ba31be3e45a130e13b278cf3b90f69bda6f6", "tree": "9bf226757e2ff84dcf73b590b3c707a7c2fba30e", "parents": [ "3301f6ae2d4cb396ae0c103329a5680d14f7a5c6", "f17936993af054b16725d0c54baa58115f9e052a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 27 11:03:24 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 27 11:03:24 2020 -0700" }, "message": "Merge tag \u0027fsnotify_for_v5.7-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs\n\nPull fanotify FAN_DIR_MODIFY disabling from Jan Kara:\n \"A single patch that disables FAN_DIR_MODIFY support that was merged in\n this merge window.\n\n When discussing further functionality we realized it may be more\n logical to guard it with a feature flag or to call things slightly\n differently (or maybe not) so let\u0027s not set the API in stone for now.\"\n\n* tag \u0027fsnotify_for_v5.7-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:\n fanotify: turn off support for FAN_DIR_MODIFY\n" }, { "commit": "3301f6ae2d4cb396ae0c103329a5680d14f7a5c6", "tree": "83b63cc5c0063392de969fe3d05b07f53da0c0d8", "parents": [ "006f38a1c3dcbe237a75e725fe457bd59cb489c4", "eec8fd0277e37cf447b88c6be181e81df867bcf1" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 27 10:58:19 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 27 10:58:19 2020 -0700" }, "message": "Merge branch \u0027for-5.7-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup\n\nPull cgroup fixes from Tejun Heo:\n\n - Reverted stricter synchronization for cgroup recursive stats which\n was prepping it for event counter usage which never got merged. The\n change was causing performation regressions in some cases.\n\n - Restore bpf-based device-cgroup operation even when cgroup1 device\n cgroup is disabled.\n\n - An out-param init fix.\n\n* \u0027for-5.7-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:\n device_cgroup: Cleanup cgroup eBPF device filter code\n xattr: fix uninitialized out-param\n Revert \"cgroup: Add memory barriers to plug cgroup_rstat_updated() race window\"\n" }, { "commit": "f17936993af054b16725d0c54baa58115f9e052a", "tree": "8c0900fd6d54e2e5482eddc9e79d9a236e3dc7be", "parents": [ "444fc5cde64330661bf59944c43844e7d4c2ccd8" ], "author": { "name": "Amir Goldstein", "email": "amir73il@gmail.com", "time": "Wed May 27 15:54:55 2020 +0300" }, "committer": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed May 27 18:55:54 2020 +0200" }, "message": "fanotify: turn off support for FAN_DIR_MODIFY\n\nFAN_DIR_MODIFY has been enabled by commit 44d705b0370b (\"fanotify:\nreport name info for FAN_DIR_MODIFY event\") in 5.7-rc1. Now we are\nplanning further extensions to the fanotify API and during that we\nrealized that FAN_DIR_MODIFY may behave slightly differently to be more\nconsistent with extensions we plan. So until we finalize these\nextensions, let\u0027s not bind our hands with exposing FAN_DIR_MODIFY to\nuserland.\n\nSigned-off-by: Amir Goldstein \u003camir73il@gmail.com\u003e\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\n" }, { "commit": "006f38a1c3dcbe237a75e725fe457bd59cb489c4", "tree": "c7dafb3ced3f45c0c3858ee22403a5d147606a99", "parents": [ "444fc5cde64330661bf59944c43844e7d4c2ccd8", "a4ae32c71fe90794127b32d26d7ad795813b502e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 27 09:53:25 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 27 09:53:25 2020 -0700" }, "message": "Merge branch \u0027exec-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace\n\nPull execve fix from Eric Biederman:\n \"While working on my exec cleanups I found a bug in exec that winds up\n miscomputing the ambient credentials during exec. Andy appears to have\n to been confused as to why credentials are computed for both the\n script and the interpreter\n\n From the original patch description:\n\n [3] Linux very confusingly processes both the script and the\n interpreter if applicable, for reasons that elude me. The results\n from thinking about a script\u0027s file capabilities and/or setuid\n bits are mostly discarded.\n\n The only value in struct cred that gets changed in cap_bprm_set_creds\n that I could find that might persist between the script and the\n interpreter was cap_ambient. Which is fixed with this trivial change\"\n\n* \u0027exec-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:\n exec: Always set cap_ambient in cap_bprm_set_creds\n" }, { "commit": "a4ae32c71fe90794127b32d26d7ad795813b502e", "tree": "c4fafbc9f1123b48542ac6c22246928e6a15aed3", "parents": [ "6a8b55ed4056ea5559ebe4f6a4b247f627870d4c" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Mon May 25 12:56:15 2020 -0500" }, "committer": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Tue May 26 13:11:00 2020 -0500" }, "message": "exec: Always set cap_ambient in cap_bprm_set_creds\n\nAn invariant of cap_bprm_set_creds is that every field in the new cred\nstructure that cap_bprm_set_creds might set, needs to be set every\ntime to ensure the fields does not get a stale value.\n\nThe field cap_ambient is not set every time cap_bprm_set_creds is\ncalled, which means that if there is a suid or sgid script with an\ninterpreter that has neither the suid nor the sgid bits set the\ninterpreter should be able to accept ambient credentials.\nUnfortuantely because cap_ambient is not reset to it\u0027s original value\nthe interpreter can not accept ambient credentials.\n\nGiven that the ambient capability set is expected to be controlled by\nthe caller, I don\u0027t think this is particularly serious. But it is\ndefinitely worth fixing so the code works correctly.\n\nI have tested to verify my reading of the code is correct and the\ninterpreter of a sgid can receive ambient capabilities with this\nchange and cannot receive ambient capabilities without this change.\n\nCc: stable@vger.kernel.org\nCc: Andy Lutomirski \u003cluto@kernel.org\u003e\nFixes: 58319057b784 (\"capabilities: ambient capabilities\")\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n" }, { "commit": "444fc5cde64330661bf59944c43844e7d4c2ccd8", "tree": "a9a700061b4ff4dadf5bcdcb48d3e6adc9cf8a1e", "parents": [ "9cb1fd0efd195590b828b9b865421ad345a4a145", "3866f217aaa81bf7165c7f27362eee5d7919c496" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 25 17:11:39 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon May 25 17:11:39 2020 -0700" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.armlinux.org.uk/~rmk/linux-arm\n\nPull ARM fixes from Russell King:\n\n - correct value of decompressor tag size in header\n\n - fix DACR value when we have nested exceptions\n\n - fix a missing newline on a kernel message\n\n - fix mask for ptrace thumb breakpoint hook\n\n* tag \u0027for-linus\u0027 of git://git.armlinux.org.uk/~rmk/linux-arm:\n ARM: 8977/1: ptrace: Fix mask for thumb breakpoint hook\n ARM: 8973/1: Add missing newline terminator to kernel message\n ARM: uaccess: fix DACR mismatch with nested exceptions\n ARM: uaccess: integrate uaccess_save and uaccess_restore\n ARM: uaccess: consolidate uaccess asm to asm/uaccess-asm.h\n ARM: 8970/1: decompressor: increase tag size\n" }, { "commit": "9cb1fd0efd195590b828b9b865421ad345a4a145", "tree": "5a800e3daf884d1b41207236613a7a1c4462c5ef", "parents": [ "98790bbac4db1697212ce9462ec35ca09c4a2810" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 24 15:32:54 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 24 15:32:54 2020 -0700" }, "message": "Linux 5.7-rc7\n" }, { "commit": "98790bbac4db1697212ce9462ec35ca09c4a2810", "tree": "608feeaa01023a0d50e687f9576aad51681e98ba", "parents": [ "667b6249b71c30c20fcd2f312f1665d90348635a", "9bb4cbf4862dfa139f73e12912129e5b76baea1a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 24 10:24:10 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 24 10:24:10 2020 -0700" }, "message": "Merge tag \u0027efi-urgent-2020-05-24\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull EFI fixes from Thomas Gleixner:\n \"A set of EFI fixes:\n\n - Don\u0027t return a garbage screen info when EFI framebuffer is not\n available\n\n - Make the early EFI console work properly with wider fonts instead\n of drawing garbage\n\n - Prevent a memory buffer leak in allocate_e820()\n\n - Print the firmware error record properly so it can be decoded by\n users\n\n - Fix a symbol clash in the host tool build which only happens with\n newer compilers.\n\n - Add a missing check for the event log version of TPM which caused\n boot failures on several Dell systems due to an attempt to decode\n SHA-1 format with the crypto agile algorithm\"\n\n* tag \u0027efi-urgent-2020-05-24\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n tpm: check event log version before reading final events\n efi: Pull up arch-specific prototype efi_systab_show_arch()\n x86/boot: Mark global variables as static\n efi: cper: Add support for printing Firmware Error Record Reference\n efi/libstub/x86: Avoid EFI map buffer alloc in allocate_e820()\n efi/earlycon: Fix early printk for wider fonts\n efi/libstub: Avoid returning uninitialized data from setup_graphics()\n" }, { "commit": "667b6249b71c30c20fcd2f312f1665d90348635a", "tree": "72547565f82eafe9b7ac98841ea759b289bf1d21", "parents": [ "9e61d12bac206e5e70176734604371ac3a6e50fd", "187b96db5ca79423618dfa29a05c438c34f9e1f0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 24 10:21:02 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 24 10:21:02 2020 -0700" }, "message": "Merge tag \u0027x86-urgent-2020-05-24\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Thomas Gleixner:\n \"Two fixes for x86:\n\n - Unbreak stack dumps for inactive tasks by interpreting the special\n first frame left by __switch_to_asm() correctly.\n\n The recent change not to skip the first frame so ORC and frame\n unwinder behave in the same way caused all entries to be\n unreliable, i.e. prepended with \u0027?\u0027.\n\n - Use cpumask_available() instead of an implicit NULL check of a\n cpumask_var_t in mmio trace to prevent a Clang build warning\"\n\n* tag \u0027x86-urgent-2020-05-24\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/unwind/orc: Fix unwind_get_return_address_ptr() for inactive tasks\n x86/mmiotrace: Use cpumask_available() for cpumask_var_t variables\n" }, { "commit": "9e61d12bac206e5e70176734604371ac3a6e50fd", "tree": "3c403704a21acad0e3dfbb0bf4e9fd8b895b7bca", "parents": [ "caffb99b6929f41a69edbb5aef3a359bf45f3315", "39f23ce07b9355d05a64ae303ce20d1c4b92b957" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 24 10:14:58 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun May 24 10:14:58 2020 -0700" }, "message": "Merge tag \u0027sched-urgent-2020-05-24\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull scheduler fixes from Thomas Gleixner:\n \"A set of fixes for the scheduler:\n\n - Fix handling of throttled parents in enqueue_task_fair() completely.\n\n The recent fix overlooked a corner case where the first iteration\n terminates due to an entity already being on the runqueue which\n makes the list management incomplete and later triggers the\n assertion which checks for completeness.\n\n - Fix a similar problem in unthrottle_cfs_rq().\n\n - Show the correct uclamp values in procfs which prints the effective\n value twice instead of requested and effective\"\n\n* tag \u0027sched-urgent-2020-05-24\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n sched/fair: Fix unthrottle_cfs_rq() for leaf_cfs_rq list\n sched/debug: Fix requested task uclamp values shown in procfs\n sched/fair: Fix enqueue_task_fair() warning some more\n" }, { "commit": "caffb99b6929f41a69edbb5aef3a359bf45f3315", "tree": "5387820f7fbdffecd08481f070655e84f22034e9", "parents": [ "0cfc8a8d70dcd51db783e8e87917e02149c71458", "539d39ad0c61b35f69565a037d7586deaf6d6166" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 17:16:18 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 17:16:18 2020 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from David Miller:\n\n 1) Fix RCU warnings in ipv6 multicast router code, from Madhuparna\n Bhowmik.\n\n 2) Nexthop attributes aren\u0027t being checked properly because of\n mis-initialized iterator, from David Ahern.\n\n 3) Revert iop_idents_reserve() change as it caused performance\n regressions and was just working around what is really a UBSAN bug\n in the compiler. From Yuqi Jin.\n\n 4) Read MAC address properly from ROM in bmac driver (double iteration\n proceeds past end of address array), from Jeremy Kerr.\n\n 5) Add Microsoft Surface device IDs to r8152, from Marc Payne.\n\n 6) Prevent reference to freed SKB in __netif_receive_skb_core(), from\n Boris Sukholitko.\n\n 7) Fix ACK discard behavior in rxrpc, from David Howells.\n\n 8) Preserve flow hash across packet scrubbing in wireguard, from Jason\n A. Donenfeld.\n\n 9) Cap option length properly for SO_BINDTODEVICE in AX25, from Eric\n Dumazet.\n\n10) Fix encryption error checking in kTLS code, from Vadim Fedorenko.\n\n11) Missing BPF prog ref release in flow dissector, from Jakub Sitnicki.\n\n12) dst_cache must be used with BH disabled in tipc, from Eric Dumazet.\n\n13) Fix use after free in mlxsw driver, from Jiri Pirko.\n\n14) Order kTLS key destruction properly in mlx5 driver, from Tariq\n Toukan.\n\n15) Check devm_platform_ioremap_resource() return value properly in\n several drivers, from Tiezhu Yang.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (71 commits)\n net: smsc911x: Fix runtime PM imbalance on error\n net/mlx4_core: fix a memory leak bug.\n net: ethernet: ti: cpsw: fix ASSERT_RTNL() warning during suspend\n net: phy: mscc: fix initialization of the MACsec protocol mode\n net: stmmac: don\u0027t attach interface until resume finishes\n net: Fix return value about devm_platform_ioremap_resource()\n net/mlx5: Fix error flow in case of function_setup failure\n net/mlx5e: CT: Correctly get flow rule\n net/mlx5e: Update netdev txq on completions during closure\n net/mlx5: Annotate mutex destroy for root ns\n net/mlx5: Don\u0027t maintain a case of del_sw_func being null\n net/mlx5: Fix cleaning unmanaged flow tables\n net/mlx5: Fix memory leak in mlx5_events_init\n net/mlx5e: Fix inner tirs handling\n net/mlx5e: kTLS, Destroy key object after destroying the TIS\n net/mlx5e: Fix allowed tc redirect merged eswitch offload cases\n net/mlx5: Avoid processing commands before cmdif is ready\n net/mlx5: Fix a race when moving command interface to events mode\n net/mlx5: Add command entry handling completion\n rxrpc: Fix a memory leak in rxkad_verify_response()\n ...\n" }, { "commit": "539d39ad0c61b35f69565a037d7586deaf6d6166", "tree": "498b24188455fdda5ce346daec97b04669d096b8", "parents": [ "e3181e9a72e19bc556e5fba6c1217901e2bc20f4" ], "author": { "name": "Dinghao Liu", "email": "dinghao.liu@zju.edu.cn", "time": "Sat May 23 16:08:20 2020 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat May 23 16:52:17 2020 -0700" }, "message": "net: smsc911x: Fix runtime PM imbalance on error\n\nRemove runtime PM usage counter decrement when the\nincrement function has not been called to keep the\ncounter balanced.\n\nSigned-off-by: Dinghao Liu \u003cdinghao.liu@zju.edu.cn\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e3181e9a72e19bc556e5fba6c1217901e2bc20f4", "tree": "4dbee0b146ce6923e1c3bbfaeea83b3c3099ebfb", "parents": [ "febfd9d3c7f74063e8e630b15413ca91b567f963", "4f7400d5cbaef676e00cdffb0565bf731c6bb09e" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat May 23 16:39:45 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat May 23 16:39:45 2020 -0700" }, "message": "Merge tag \u0027mlx5-fixes-2020-05-22\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux\n\nSaeed Mahameed says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nmlx5 fixes 2020-05-22\n\nThis series introduces some fixes to mlx5 driver.\n\nPlease pull and let me know if there is any problem.\n\nFor -stable v4.13\n (\u0027net/mlx5: Add command entry handling completion\u0027)\n\nFor -stable v5.2\n (\u0027net/mlx5: Fix error flow in case of function_setup failure\u0027)\n (\u0027net/mlx5: Fix memory leak in mlx5_events_init\u0027)\n\nFor -stable v5.3\n (\u0027net/mlx5e: Update netdev txq on completions during closure\u0027)\n (\u0027net/mlx5e: kTLS, Destroy key object after destroying the TIS\u0027)\n (\u0027net/mlx5e: Fix inner tirs handling\u0027)\n\nFor -stable v5.6\n (\u0027net/mlx5: Fix cleaning unmanaged flow tables\u0027)\n (\u0027net/mlx5: Fix a race when moving command interface to events mode\u0027)\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": "febfd9d3c7f74063e8e630b15413ca91b567f963", "tree": "f639112705778933bb6d0bd07f204ab33e16b03d", "parents": [ "4c64b83d03f4aafcdf710caad994cbc855802e74" ], "author": { "name": "Qiushi Wu", "email": "wu000273@umn.edu", "time": "Fri May 22 14:07:15 2020 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat May 23 16:34:37 2020 -0700" }, "message": "net/mlx4_core: fix a memory leak bug.\n\nIn function mlx4_opreq_action(), pointer \"mailbox\" is not released,\nwhen mlx4_cmd_box() return and error, causing a memory leak bug.\nFix this issue by going to \"out\" label, mlx4_free_cmd_mailbox() can\nfree this pointer.\n\nFixes: fe6f700d6cbb (\"net/mlx4_core: Respond to operation request by firmware\")\nSigned-off-by: Qiushi Wu \u003cwu000273@umn.edu\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4c64b83d03f4aafcdf710caad994cbc855802e74", "tree": "14cbc0f0748fbac0c44befa5a1764cc311dca2a5", "parents": [ "0ddfee1feece1c85592d49b759286032ef2dd803" ], "author": { "name": "Grygorii Strashko", "email": "grygorii.strashko@ti.com", "time": "Fri May 22 20:09:28 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat May 23 16:33:20 2020 -0700" }, "message": "net: ethernet: ti: cpsw: fix ASSERT_RTNL() warning during suspend\n\nvlan_for_each() are required to be called with rtnl_lock taken, otherwise\nASSERT_RTNL() warning will be triggered - which happens now during System\nresume from suspend:\n cpsw_suspend()\n |- cpsw_ndo_stop()\n |- __hw_addr_ref_unsync_dev()\n |- cpsw_purge_all_mc()\n |- vlan_for_each()\n |- ASSERT_RTNL();\n\nHence, fix it by surrounding cpsw_ndo_stop() by rtnl_lock/unlock() calls.\n\nFixes: 15180eca569b (\"net: ethernet: ti: cpsw: fix vlan mcast\")\nSigned-off-by: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0ddfee1feece1c85592d49b759286032ef2dd803", "tree": "6792d99bf9b7a103202a4eaec791809eef4bd794", "parents": [ "31096c3e8b1163c6e966bf4d1f36d8b699008f84" ], "author": { "name": "Antoine Tenart", "email": "antoine.tenart@bootlin.com", "time": "Fri May 22 17:55:45 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat May 23 16:31:53 2020 -0700" }, "message": "net: phy: mscc: fix initialization of the MACsec protocol mode\n\nAt the very end of the MACsec block initialization in the MSCC PHY\ndriver, the MACsec \"protocol mode\" is set. This setting should be set\nbased on the PHY id within the package, as the bank used to access the\nregister used depends on this. This was not done correctly, and only the\nfirst bank was used leading to the two upper PHYs being unstable when\nusing the VSC8584. This patch fixes it.\n\nFixes: 1bbe0ecc2a1a (\"net: phy: mscc: macsec initialization\")\nSigned-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "31096c3e8b1163c6e966bf4d1f36d8b699008f84", "tree": "4d556908b3e0b632161db17ae2bbccd8a488060d", "parents": [ "ef24d6c3d6965158dfe23ae961d87e9a343e18a2" ], "author": { "name": "Leon Yu", "email": "leoyu@nvidia.com", "time": "Fri May 22 23:29:43 2020 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat May 23 16:30:27 2020 -0700" }, "message": "net: stmmac: don\u0027t attach interface until resume finishes\n\nCommit 14b41a2959fb (\"net: stmmac: Delete txtimer in suspend\") was the\nfirst attempt to fix a race between mod_timer() and setup_timer()\nduring stmmac_resume(). However the issue still exists as the commit\nonly addressed half of the issue.\n\nSame race can still happen as stmmac_resume() re-attaches interface\nway too early - even before hardware is fully initialized. Worse,\ndoing so allows network traffic to restart and stmmac_tx_timer_arm()\nbeing called in the middle of stmmac_resume(), which re-init tx timers\nin stmmac_init_coalesce(). timer_list will be corrupted and system\ncrashes as a result of race between mod_timer() and setup_timer().\n\n systemd--1995 2.... 552950018us : stmmac_suspend: 4994\n ksoftirq-9 0..s2 553123133us : stmmac_tx_timer_arm: 2276\n systemd--1995 0.... 553127896us : stmmac_resume: 5101\n systemd--320 7...2 553132752us : stmmac_tx_timer_arm: 2276\n (sd-exec-1999 5...2 553135204us : stmmac_tx_timer_arm: 2276\n ---------------------------------\n pc : run_timer_softirq+0x468/0x5e0\n lr : run_timer_softirq+0x570/0x5e0\n Call trace:\n run_timer_softirq+0x468/0x5e0\n __do_softirq+0x124/0x398\n irq_exit+0xd8/0xe0\n __handle_domain_irq+0x6c/0xc0\n gic_handle_irq+0x60/0xb0\n el1_irq+0xb8/0x180\n arch_cpu_idle+0x38/0x230\n default_idle_call+0x24/0x3c\n do_idle+0x1e0/0x2b8\n cpu_startup_entry+0x28/0x48\n secondary_start_kernel+0x1b4/0x208\n\nFix this by deferring netif_device_attach() to the end of\nstmmac_resume().\n\nSigned-off-by: Leon Yu \u003cleoyu@nvidia.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ef24d6c3d6965158dfe23ae961d87e9a343e18a2", "tree": "923489ccb4df091a557366098084e2541dc0a713", "parents": [ "d04322a0da1e86aedaa322ce933cfb8c0191d1eb" ], "author": { "name": "Tiezhu Yang", "email": "yangtiezhu@loongson.cn", "time": "Fri May 22 19:03:21 2020 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat May 23 16:28:25 2020 -0700" }, "message": "net: Fix return value about devm_platform_ioremap_resource()\n\nWhen call function devm_platform_ioremap_resource(), we should use IS_ERR()\nto check the return value and return PTR_ERR() if failed.\n\nSigned-off-by: Tiezhu Yang \u003cyangtiezhu@loongson.cn\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0cfc8a8d70dcd51db783e8e87917e02149c71458", "tree": "68b7293e8da1f1cdee1f5fb98d44f7ff9453b1c6", "parents": [ "423b8baf18a8c03f2d6fa99aa447ed0da189bb95" ], "author": { "name": "Mike Rapoport", "email": "rppt@linux.ibm.com", "time": "Sat May 23 22:57:18 2020 +0300" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 13:59:12 2020 -0700" }, "message": "sparc32: fix page table traversal in srmmu_nocache_init()\n\nThe srmmu_nocache_init() uses __nocache_fix() macro to add an offset to\npage table entry to access srmmu_nocache_pool.\n\nBut since sparc32 has only three actual page table levels, pgd, p4d and\npud are essentially the same thing and pgd_offset() and p4d_offset() are\nno-ops, the __nocache_fix() should be done only at PUD level.\n\nRemove __nocache_fix() for p4d_offset() and pud_offset() and keep it\nonly for PUD and lower levels.\n\nFixes: c2bc26f7ca1f (\"sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()\")\nSigned-off-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: David S. Miller \u003cdavem@davemloft.net\u003e\nCc: Anatoly Pugachev \u003cmatorola@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "423b8baf18a8c03f2d6fa99aa447ed0da189bb95", "tree": "68d166214b4cde3da55a1159df6fdab6a4a4a568", "parents": [ "23f0dac848412dafd197566b62d831d5a68b5b6b", "ca6edee6c6e3fe4ec143ee1964593d755ddbee62" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 11:21:47 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 11:21:47 2020 -0700" }, "message": "Merge branch \u0027akpm\u0027 (patches from Andrew)\n\nMerge misc fixes from Andrew Morton:\n \"11 fixes\"\n\n* emailed patches from Andrew Morton \u003cakpm@linux-foundation.org\u003e:\n MAINTAINERS: add files related to kdump\n z3fold: fix use-after-free when freeing handles\n sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()\n MAINTAINERS: update email address for Naoya Horiguchi\n sh: include linux/time_types.h for sockios\n kasan: disable branch tracing for core runtime\n selftests/vm/write_to_hugetlbfs.c: fix unused variable warning\n selftests/vm/.gitignore: add mremap_dontunmap\n rapidio: fix an error in get_user_pages_fast() error handling\n x86: bitops: fix build regression\n device-dax: don\u0027t leak kernel memory to user space after unloading kmem\n" }, { "commit": "23f0dac848412dafd197566b62d831d5a68b5b6b", "tree": "9b7fe649f902d0b40f42e8f0c32a55a954b1a590", "parents": [ "0e36fd45e37e0b4151d70cfae4d7e3e9a75c5c99", "e6764aa0e5530066dd969eccea2a1a7d177859a8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 11:06:13 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 11:06:13 2020 -0700" }, "message": "Merge tag \u0027driver-core-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core\n\nPull driver core fixes from Greg KH:\n \"So, turns out the kobject fix didn\u0027t quite work, so here are four\n patches that in the end, result in just two driver core fixes for\n reported issues that no one has had problems with.\n\n The kobject patch that was originally in here has now been reverted,\n as Guenter reported boot problems with it on some of his systems\"\n\n* tag \u0027driver-core-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:\n Revert \"kobject: Make sure the parent does not get released before its children\"\n kobject: Make sure the parent does not get released before its children\n driver core: Fix handling of SYNC_STATE_ONLY + STATELESS device links\n driver core: Fix SYNC_STATE_ONLY device link implementation\n" }, { "commit": "0e36fd45e37e0b4151d70cfae4d7e3e9a75c5c99", "tree": "370acba9beb8873ae261a7e25307bc6719678c31", "parents": [ "8f261041b18ee80ad8afdd1621c909c4df9f6cc3", "7a839dbab1be59f5ed3b3b046de29e166784c9b4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 11:02:42 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 11:02:42 2020 -0700" }, "message": "Merge tag \u0027char-misc-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc\n\nPull char/misc fixes from Greg KH:\n \"Here are some small char/misc driver fixes for 5.7-rc7 that resolve\n some reported issues. Included in here are tiny fixes for the mei,\n coresight, rtsx, ipack, and mhi drivers.\n\n All of these have been in linux-next with no reported issues\"\n\n* tag \u0027char-misc-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:\n misc: rtsx: Add short delay after exit from ASPM\n bus: mhi: core: Fix some error return code\n ipack: tpci200: fix error return code in tpci200_register()\n coresight: cti: remove incorrect NULL return check\n mei: release me_cl object reference\n" }, { "commit": "8f261041b18ee80ad8afdd1621c909c4df9f6cc3", "tree": "aff4d89de8c61e58163b857254454c81ad4b1345", "parents": [ "d3044d7d220529a4ce0cbfe00de63996935908d1", "bcb392871813ef9e233645557d6dd85bb1a6f41a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:57:55 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:57:55 2020 -0700" }, "message": "Merge tag \u0027staging-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging\n\nPull staging/iio fixes from Greg KH:\n \"Here are some small staging and IIO driver fixes for 5.7-rc7\n\n Nothing major, just a collection of IIO driver fixes for reported\n issues, and a few small staging driver fixes that people have found.\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 \u0027staging-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:\n staging: wfx: unlock on error path\n staging: greybus: Fix uninitialized scalar variable\n staging: kpc2000: fix error return code in kp2000_pcie_probe()\n iio: sca3000: Remove an erroneous \u0027get_device()\u0027\n iio: adc: stm32-dfsdm: fix device used to request dma\n iio: adc: stm32-adc: fix device used to request dma\n iio: adc: ti-ads8344: Fix channel selection\n staging: iio: ad2s1210: Fix SPI reading\n iio: dac: vf610: Fix an error handling path in \u0027vf610_dac_probe()\u0027\n iio: imu: st_lsm6dsx: unlock on error in st_lsm6dsx_shub_write_raw()\n iio: chemical: atlas-sensor: correct DO-SM channels\n" }, { "commit": "d3044d7d220529a4ce0cbfe00de63996935908d1", "tree": "0db6a145ba7d2946f598030d85e01472a7642186", "parents": [ "9bca7c40850eca2613d79878e6d730faeaa602cf", "17b4efdf4e4867079012a48ca10d965fe9d68822" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:50:38 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:50:38 2020 -0700" }, "message": "Merge tag \u0027tty-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty\n\nPull tty/serial fix from Greg KH:\n \"Here is a single serial driver fix for 5.7-rc7. It resolves an issue\n with the SiFive serial console init sequence that was reported a\n number of times.\n\n It has been in linux-next for a while now with no reported issues\"\n\n* tag \u0027tty-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:\n tty: serial: add missing spin_lock_init for SiFive serial console\n" }, { "commit": "9bca7c40850eca2613d79878e6d730faeaa602cf", "tree": "12d6c145a1b194bb3bb028f4bddb34573440934b", "parents": [ "e644645abf4788e919beeb97925fb6bf43e890a2", "4c1cbcbd6c56c79de2c07159be4f55386bb0bef2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:42:12 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:42:12 2020 -0700" }, "message": "Merge tag \u0027s390-5.7-4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\nPull s390 fixes from Vasily Gorbik:\n\n - Add missing R_390_JMP_SLOT relocation type in KASLR code.\n\n - Fix set_huge_pte_at for empty ptes issue which has been uncovered\n with arch page table helper tests.\n\n - Correct initrd location for kdump kernel.\n\n - Fix s390_mmio_read/write with MIO in PCI code.\n\n* tag \u0027s390-5.7-4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:\n s390/kaslr: add support for R_390_JMP_SLOT relocation type\n s390/mm: fix set_huge_pte_at() for empty ptes\n s390/kexec_file: fix initrd location for kdump kernel\n s390/pci: Fix s390_mmio_read/write with MIO\n" }, { "commit": "ca6edee6c6e3fe4ec143ee1964593d755ddbee62", "tree": "17cc47e874d61fc1f402c23dad834293d7f990cc", "parents": [ "d8f117abb380ba968b5e3ef2042d901c02872a4c" ], "author": { "name": "Baoquan He", "email": "bhe@redhat.com", "time": "Fri May 22 22:23:15 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:32 2020 -0700" }, "message": "MAINTAINERS: add files related to kdump\n\nKdump is implemented based on kexec, however some files are only related\nto crash dumping and missing, add them to KDUMP entry.\n\nSigned-off-by: Baoquan He \u003cbhe@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nAcked-by: Dave Young \u003cdyoung@redhat.com\u003e\nLink: http://lkml.kernel.org/r/20200520103633.GW5029@MiWiFi-R3L-srv\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d8f117abb380ba968b5e3ef2042d901c02872a4c", "tree": "8d38eb4409e79bf06db06520e0063dd57038d002", "parents": [ "c2bc26f7ca1ff1165bb6669a7a4cccc20ffd2ced" ], "author": { "name": "Uladzislau Rezki", "email": "uladzislau.rezki@sony.com", "time": "Fri May 22 22:23:12 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:32 2020 -0700" }, "message": "z3fold: fix use-after-free when freeing handles\n\nfree_handle() for a foreign handle may race with inter-page compaction,\nwhat can lead to memory corruption.\n\nTo avoid that, take write lock not read lock in free_handle to be\nsynchronized with __release_z3fold_page().\n\nFor example KASAN can detect it:\n\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n BUG: KASAN: use-after-free in LZ4_decompress_safe+0x2c4/0x3b8\n Read of size 1 at addr ffffffc976695ca3 by task GoogleApiHandle/4121\n\n CPU: 0 PID: 4121 Comm: GoogleApiHandle Tainted: P S OE 4.19.81-perf+ #162\n Hardware name: Sony Mobile Communications. PDX-203(KONA) (DT)\n Call trace:\n LZ4_decompress_safe+0x2c4/0x3b8\n lz4_decompress_crypto+0x3c/0x70\n crypto_decompress+0x58/0x70\n zcomp_decompress+0xd4/0x120\n ...\n\nApart from that, initialize zhdr-\u003emapped_count in init_z3fold_page() and\nremove \"newpage\" variable because it is not used anywhere.\n\nSigned-off-by: Uladzislau Rezki \u003culadzislau.rezki@sony.com\u003e\nSigned-off-by: Vitaly Wool \u003cvitaly.wool@konsulko.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Qian Cai \u003ccai@lca.pw\u003e\nCc: Raymond Jennings \u003cshentino@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/20200520082100.28876-1-vitaly.wool@konsulko.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c2bc26f7ca1ff1165bb6669a7a4cccc20ffd2ced", "tree": "c8fe6b2a13ecae9940f6f458099cf10af8812975", "parents": [ "f7fa1876af81512444631d324adb77383f56c37a" ], "author": { "name": "Mike Rapoport", "email": "rppt@linux.ibm.com", "time": "Fri May 22 22:23:09 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:32 2020 -0700" }, "message": "sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()\n\nThe kbuild test robot reported the following warning:\n\n arch/sparc/mm/srmmu.c: In function \u0027srmmu_nocache_init\u0027: arch/sparc/mm/srmmu.c:300:9: error: variable \u0027pud\u0027 set but not used [-Werror\u003dunused-but-set-variable]\n 300 | pud_t *pud;\n\nThis warning is caused by misprint in the page table traversal in\nsrmmu_nocache_init() function which accessed a PMD entry using PGD\nrather than PUD.\n\nSince sparc32 has only 3 page table levels, the PGD and PUD are\nessentially the same and usage of __nocache_fix() removed the type\nchecking.\n\nUse PUD for the consistency and to silence the compiler warning.\n\nFixes: 7235db268a2777bc38 (\"sparc32: use pgtable-nopud instead of 4level-fixup\")\nReported-by: kbuild test robot \u003clkp@intel.com\u003e\nSigned-off-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: David S. Miller \u003cdavem@davemloft.net\u003e\nCc: Anatoly Pugachev \u003cmatorola@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/20200520132005.GM1059226@linux.ibm.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f7fa1876af81512444631d324adb77383f56c37a", "tree": "14337f16b13b28139a20b0cee1e23e8b22d4c030", "parents": [ "fc94cf2092c7c1267fa2deb8388d624f50eba808" ], "author": { "name": "Naoya Horiguchi", "email": "n-horiguchi@ah.jp.nec.com", "time": "Fri May 22 22:23:06 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:31 2020 -0700" }, "message": "MAINTAINERS: update email address for Naoya Horiguchi\n\nMy email address has changed due to system upgrade, so please update it\nin MAINTAINERS list. My old address (n-horiguchi@ah.jp.nec.com) will be\nstill active for a few months.\n\nNote that my email system has some encoding issue and can\u0027t send patches\nin raw format via git-send-email. So patches from me will be delivered\nvia my free address (nao.horiguchi@gmail.com) or GitHub.\n\nSigned-off-by: Naoya Horiguchi \u003cnaoya.horiguchi@nec.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nLink: http://lkml.kernel.org/r/1589874488-9247-1-git-send-email-naoya.horiguchi@nec.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fc94cf2092c7c1267fa2deb8388d624f50eba808", "tree": "08f49f8c08268a3f3bc638939c8cab987dad8030", "parents": [ "33cd65e73abd693c00c4156cf23677c453b41b3b" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Fri May 22 22:23:02 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:31 2020 -0700" }, "message": "sh: include linux/time_types.h for sockios\n\nUsing the socket ioctls on arch/sh (and only there) causes build time\nproblems when __kernel_old_timeval/__kernel_old_timespec are not already\nvisible to the compiler.\n\nAdd an explict include line for the header that defines these\nstructures.\n\nFixes: 8c709f9a0693 (\"y2038: sh: remove timeval/timespec usage from headers\")\nFixes: 0768e17073dc (\"net: socket: implement 64-bit timestamps\")\nReported-by: John Paul Adrian Glaubitz \u003cglaubitz@physik.fu-berlin.de\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nTested-by: John Paul Adrian Glaubitz \u003cglaubitz@physik.fu-berlin.de\u003e\nCc: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\nCc: Rich Felker \u003cdalias@libc.org\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: John Paul Adrian Glaubitz \u003cglaubitz@physik.fu-berlin.de\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/20200519131327.1836482-1-arnd@arndb.de\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "33cd65e73abd693c00c4156cf23677c453b41b3b", "tree": "cf1237060c2eb28ccf1a3916d1bf0aea5fdb6654", "parents": [ "380e5c1d9b5e9768315eb4d405c127cb655406b3" ], "author": { "name": "Marco Elver", "email": "elver@google.com", "time": "Fri May 22 22:22:59 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:31 2020 -0700" }, "message": "kasan: disable branch tracing for core runtime\n\nDuring early boot, while KASAN is not yet initialized, it is possible to\nenter reporting code-path and end up in kasan_report().\n\nWhile uninitialized, the branch there prevents generating any reports,\nhowever, under certain circumstances when branches are being traced\n(TRACE_BRANCH_PROFILING), we may recurse deep enough to cause kernel\nreboots without warning.\n\nTo prevent similar issues in future, we should disable branch tracing\nfor the core runtime.\n\n[elver@google.com: remove duplicate DISABLE_BRANCH_PROFILING, per Qian Cai]\n Link: https://lore.kernel.org/lkml/20200517011732.GE24705@shao2-debian/\n Link: http://lkml.kernel.org/r/20200522075207.157349-1-elver@google.com\nReported-by: kernel test robot \u003crong.a.chen@intel.com\u003e\nSigned-off-by: Marco Elver \u003celver@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nCc: Dmitry Vyukov \u003cdvyukov@google.com\u003e\nCc: Alexander Potapenko \u003cglider@google.com\u003e\nCc: Andrey Ryabinin \u003caryabinin@virtuozzo.com\u003e\nCc: Qian Cai \u003ccai@lca.pw\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r//20200517011732.GE24705@shao2-debian/\nLink: http://lkml.kernel.org/r/20200519182459.87166-1-elver@google.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "380e5c1d9b5e9768315eb4d405c127cb655406b3", "tree": "d6bd374ebe930a0f70eee1ee6157f3583cf7a997", "parents": [ "98097701cc0bec06e4bc183cceaf6dfa06a69e10" ], "author": { "name": "John Hubbard", "email": "jhubbard@nvidia.com", "time": "Fri May 22 22:22:56 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:31 2020 -0700" }, "message": "selftests/vm/write_to_hugetlbfs.c: fix unused variable warning\n\nRemove unused variable \"i\", which was triggering a compiler warning.\n\nFixes: 29750f71a9b4 (\"hugetlb_cgroup: add hugetlb_cgroup reservation tests\")\nSigned-off-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-By: Mina Almasry \u003calmasrymina@google.com\u003e\nCc: Brian Geffon \u003cbgeffon@google.com\u003e\nCc: \"Kirill A . Shutemov\" \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Shuah Khan \u003cshuah@kernel.org\u003e\nCc: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nLink: http://lkml.kernel.org/r/20200517001245.361762-2-jhubbard@nvidia.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "98097701cc0bec06e4bc183cceaf6dfa06a69e10", "tree": "cf3e6ba28c1f6bd7ebcd91920d65f3222243f5b1", "parents": [ "ffca476a0a8d26de767cc41d62b8ca7f540ecfdd" ], "author": { "name": "John Hubbard", "email": "jhubbard@nvidia.com", "time": "Fri May 22 22:22:53 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:31 2020 -0700" }, "message": "selftests/vm/.gitignore: add mremap_dontunmap\n\nAdd mremap_dontunmap to .gitignore.\n\nFixes: 0c28759ee3c9 (\"selftests: add MREMAP_DONTUNMAP selftest\")\nSigned-off-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Brian Geffon \u003cbgeffon@google.com\u003e\nLink: http://lkml.kernel.org/r/20200517002509.362401-2-jhubbard@nvidia.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ffca476a0a8d26de767cc41d62b8ca7f540ecfdd", "tree": "44203123925dc974633d5cf08818186e636dc3e5", "parents": [ "c071b0f11e7fb944525b12c80e728af69648d967" ], "author": { "name": "John Hubbard", "email": "jhubbard@nvidia.com", "time": "Fri May 22 22:22:48 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:31 2020 -0700" }, "message": "rapidio: fix an error in get_user_pages_fast() error handling\n\nIn the case of get_user_pages_fast() returning fewer pages than\nrequested, rio_dma_transfer() does not quite do the right thing. It\nattempts to release all the pages that were requested, rather than just\nthe pages that were pinned.\n\nFix the error handling so that only the pages that were successfully\npinned are released.\n\nFixes: e8de370188d0 (\"rapidio: add mport char device driver\")\nSigned-off-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Matt Porter \u003cmporter@kernel.crashing.org\u003e\nCc: Alexandre Bounine \u003calex.bou9@gmail.com\u003e\nCc: Sumit Semwal \u003csumit.semwal@linaro.org\u003e\nCc: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/20200517235620.205225-2-jhubbard@nvidia.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c071b0f11e7fb944525b12c80e728af69648d967", "tree": "add23517de24537d3c81f75a37a2b49eec60a7f1", "parents": [ "60858c00e5f018eda711a3aa84cf62214ef62d61" ], "author": { "name": "Nick Desaulniers", "email": "ndesaulniers@google.com", "time": "Fri May 22 22:22:45 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:31 2020 -0700" }, "message": "x86: bitops: fix build regression\n\nThis is easily reproducible via CC\u003dclang + CONFIG_STAGING\u003dy +\nCONFIG_VT6656\u003dm.\n\nIt turns out that if your config tickles __builtin_constant_p via\ndifferences in choices to inline or not, these statements produce\ninvalid assembly:\n\n $ cat foo.c\n long a(long b, long c) {\n asm(\"orb\t%1, %0\" : \"+q\"(c): \"r\"(b));\n return c;\n }\n $ gcc foo.c\n foo.c: Assembler messages:\n foo.c:2: Error: `%rax\u0027 not allowed with `orb\u0027\n\nUse the `%b` \"x86 Operand Modifier\" to instead force register allocation\nto select a lower-8-bit GPR operand.\n\nThe \"q\" constraint only has meaning on -m32 otherwise is treated as\n\"r\". Not all GPRs have low-8-bit aliases for -m32.\n\nFixes: 1651e700664b4 (\"x86: Fix bitops.h warning with a moved cast\")\nReported-by: kernelci.org bot \u003cbot@kernelci.org\u003e\nSuggested-by: Andy Shevchenko \u003candriy.shevchenko@intel.com\u003e\nSuggested-by: Brian Gerst \u003cbrgerst@gmail.com\u003e\nSuggested-by: H. Peter Anvin \u003chpa@zytor.com\u003e\nSuggested-by: Ilie Halip \u003cilie.halip@gmail.com\u003e\nSigned-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nTested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\nTested-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\t[build, clang-11]\nReviewed-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nReviewed-By: Brian Gerst \u003cbrgerst@gmail.com\u003e\nReviewed-by: Jesse Brandeburg \u003cjesse.brandeburg@intel.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Borislav Petkov \u003cbp@alien8.de\u003e\nCc: Marco Elver \u003celver@google.com\u003e\nCc: \"Paul E. McKenney\" \u003cpaulmck@kernel.org\u003e\nCc: Andrey Ryabinin \u003caryabinin@virtuozzo.com\u003e\nCc: Luc Van Oostenryck \u003cluc.vanoostenryck@gmail.com\u003e\nCc: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nCc: Daniel Axtens \u003cdja@axtens.net\u003e\nCc: \"Peter Zijlstra (Intel)\" \u003cpeterz@infradead.org\u003e\nLink: http://lkml.kernel.org/r/20200508183230.229464-1-ndesaulniers@google.com\nLink: https://github.com/ClangBuiltLinux/linux/issues/961\nLink: https://lore.kernel.org/lkml/20200504193524.GA221287@google.com/\nLink: https://gcc.gnu.org/onlinedocs/gcc/Extended-Asm.html#x86Operandmodifiers\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "60858c00e5f018eda711a3aa84cf62214ef62d61", "tree": "f52a83f811cad052f669497af7e0259170ca3295", "parents": [ "444565650a5fe9c63ddf153e6198e31705dedeb2" ], "author": { "name": "David Hildenbrand", "email": "david@redhat.com", "time": "Fri May 22 22:22:42 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 23 10:26:31 2020 -0700" }, "message": "device-dax: don\u0027t leak kernel memory to user space after unloading kmem\n\nAssume we have kmem configured and loaded:\n\n [root@localhost ~]# cat /proc/iomem\n ...\n 140000000-33fffffff : Persistent Memory$\n 140000000-1481fffff : namespace0.0\n 150000000-33fffffff : dax0.0\n 150000000-33fffffff : System RAM\n\nAssume we try to unload kmem. This force-unloading will work, even if\nmemory cannot get removed from the system.\n\n [root@localhost ~]# rmmod kmem\n [ 86.380228] removing memory fails, because memory [0x0000000150000000-0x0000000157ffffff] is onlined\n ...\n [ 86.431225] kmem dax0.0: DAX region [mem 0x150000000-0x33fffffff] cannot be hotremoved until the next reboot\n\nNow, we can reconfigure the namespace:\n\n [root@localhost ~]# ndctl create-namespace --force --reconfig\u003dnamespace0.0 --mode\u003ddevdax\n [ 131.409351] nd_pmem namespace0.0: could not reserve region [mem 0x140000000-0x33fffffff]dax\n [ 131.410147] nd_pmem: probe of namespace0.0 failed with error -16namespace0.0 --mode\u003ddevdax\n ...\n\nThis fails as expected due to the busy memory resource, and the memory\ncannot be used. However, the dax0.0 device is removed, and along its\nname.\n\nThe name of the memory resource now points at freed memory (name of the\ndevice):\n\n [root@localhost ~]# cat /proc/iomem\n ...\n 140000000-33fffffff : Persistent Memory\n 140000000-1481fffff : namespace0.0\n 150000000-33fffffff : �_�^7_��/_��wR��WQ���^��� ...\n 150000000-33fffffff : System RAM\n\nWe have to make sure to duplicate the string. While at it, remove the\nsuperfluous setting of the name and fixup a stale comment.\n\nFixes: 9f960da72b25 (\"device-dax: \"Hotremove\" persistent memory that is used like normal RAM\")\nSigned-off-by: David Hildenbrand \u003cdavid@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Dan Williams \u003cdan.j.williams@intel.com\u003e\nCc: Vishal Verma \u003cvishal.l.verma@intel.com\u003e\nCc: Dave Jiang \u003cdave.jiang@intel.com\u003e\nCc: Pavel Tatashin \u003cpasha.tatashin@soleen.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: \u003cstable@vger.kernel.org\u003e\t[5.3]\nLink: http://lkml.kernel.org/r/20200508084217.9160-2-david@redhat.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e6764aa0e5530066dd969eccea2a1a7d177859a8", "tree": "5d29a474bd1e410c685e27b81bf11cf38b244c0b", "parents": [ "4ef12f7198023c09ad6d25b652bd8748c965c7fa" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sat May 23 17:11:11 2020 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Sat May 23 17:11:11 2020 +0200" }, "message": "Revert \"kobject: Make sure the parent does not get released before its children\"\n\nThis reverts commit 4ef12f7198023c09ad6d25b652bd8748c965c7fa.\n\nGuenter reports:\n\n\tAll my arm64be (arm64 big endian) boot tests crash with this\n\tpatch applied. Reverting it fixes the problem. Crash log and\n\tbisect results (from pending-fixes branch) below.\n\nAnd also:\n\tarm64 images don\u0027t crash but report lots of \"poison overwritten\"\n\tbacktraces like the one below. On arm, I see \"refcount_t:\n\tunderflow\", also attached. I didn\u0027t bisect those, but given the\n\tcontext I would suspect the same culprit.\n\nReported-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\nLink: https://lore.kernel.org/r/20200513151840.36400-1-heikki.krogerus@linux.intel.com\nCc: Naresh Kamboju \u003cnaresh.kamboju@linaro.org\u003e\nCc: kernel test robot \u003crong.a.chen@intel.com\u003e\nCc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\nCc: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nCc: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nCc: Randy Dunlap \u003crdunlap@infradead.org\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "e644645abf4788e919beeb97925fb6bf43e890a2", "tree": "3ec2e377dfe03eaabce18d5cd05fc6fe424097d6", "parents": [ "a24deb9d68b423f3e5e190b9836d192ee3e0ff9d", "f2e6b75f6ee82308ef7b00f29e71e5f1c6b3d52a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 21:43:02 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 21:43:02 2020 -0700" }, "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 \"Three minor fixes, two in drivers, one to fix a hang after reset with\n iSCSI, and one to avoid a spurious log message; and the final core one\n to correct a suspend/resume miscount with quiesced devices\"\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n scsi: target: Put lun_ref at end of tmr processing\n scsi: pm: Balance pm_only counter of request queue during system resume\n scsi: qla2xxx: Do not log message when reading port speed via sysfs\n" }, { "commit": "a24deb9d68b423f3e5e190b9836d192ee3e0ff9d", "tree": "9beaab856ddd6e64046c179f6858c67e0a282e83", "parents": [ "444565650a5fe9c63ddf153e6198e31705dedeb2", "7d9ff5eed4146bf026c69e766ff630bc0bd555bb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 21:37:09 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 21:37:09 2020 -0700" }, "message": "Merge tag \u0027drm-fixes-2020-05-23\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Things seemed to have quieten down, though no i915 pull (I even gave\n them an extra 12 hours in case they were late).\n\n The amdgpu floating point fix is probably the largest, but it just\n moves some code around to it doesn\u0027t do fpu stuff outside the fpu\n boundaries. Otherwise it\u0027s just a couple of vmwgfx fixes (maintainer\n change) and two etnaviv fixes.\n\n vmwgfx:\n - change maintainers\n - fix redundant assignment\n - fix parameter name\n - fix return value\n\n etnaviv:\n - memory leak fix when userspace passes a invalid softpin address\n - off-by-one crashing the kernel in the perfmon domain iteration when\n the GPU core has both 2D and 3D capabilities\n\n amdgpu:\n - DP fix\n - Floating point fix\n - Fix cursor stutter issue\"\n\n* tag \u0027drm-fixes-2020-05-23\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/amd/display: Defer cursor lock until after VUPDATE\n drm/amd/display: Remove dml_common_def file\n drm/amd/display: DP training to set properly SCRAMBLING_DISABLE\n drm/edid: Add Oculus Rift S to non-desktop list\n drm/etnaviv: Fix a leak in submit_pin_objects()\n drm/etnaviv: fix perfmon domain interation\n drm/vmwgfx: Return true in function vmw_fence_obj_signaled()\n drm/vmwgfx: remove redundant assignment to variable ret\n drm/vmwgfx: Fix parameter name in vmw_bo_init\n drm/vmwgfx: update MAINTAINERS entry\n" }, { "commit": "4f7400d5cbaef676e00cdffb0565bf731c6bb09e", "tree": "ed5d29c9ed8eac34652aa96e5097ea5d5daf621a", "parents": [ "d37bd5e81ed0d58f0ebe2e01658c26722e0c033e" ], "author": { "name": "Shay Drory", "email": "shayd@mellanox.com", "time": "Wed May 06 14:52:04 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:58 2020 -0700" }, "message": "net/mlx5: Fix error flow in case of function_setup failure\n\nCurrently, if an error occurred during mlx5_function_setup(), we\nkeep dev-\u003estate as DEVICE_STATE_UP.\nFixing it by adding a goto label.\n\nFixes: e161105e58da (\"net/mlx5: Function setup/teardown procedures\")\nSigned-off-by: Shay Drory \u003cshayd@mellanox.com\u003e\nReviewed-by: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "d37bd5e81ed0d58f0ebe2e01658c26722e0c033e", "tree": "3893e91ca16310f8f4e2ef6829781819bd8bd205", "parents": [ "5e911e2c06bd8c17df29147a5e2d4b17fafda024" ], "author": { "name": "Roi Dayan", "email": "roid@mellanox.com", "time": "Mon May 18 20:21:11 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:56 2020 -0700" }, "message": "net/mlx5e: CT: Correctly get flow rule\n\nThe correct way is to us the flow_cls_offload_flow_rule() wrapper\ninstead of f-\u003erule directly.\n\nFixes: 4c3844d9e97e (\"net/mlx5e: CT: Introduce connection tracking\")\nSigned-off-by: Roi Dayan \u003croid@mellanox.com\u003e\nReviewed-by: Oz Shlomo \u003cozsh@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "5e911e2c06bd8c17df29147a5e2d4b17fafda024", "tree": "89919bb328a0a509b3d6330c6c15b23b02bb616f", "parents": [ "9ca415399dae133b00273a4283ef31d003a6818d" ], "author": { "name": "Moshe Shemesh", "email": "moshe@mellanox.com", "time": "Tue Apr 07 17:38:28 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:54 2020 -0700" }, "message": "net/mlx5e: Update netdev txq on completions during closure\n\nOn sq closure when we free its descriptors, we should also update netdev\ntxq on completions which would not arrive. Otherwise if we reopen sqs\nand attach them back, for example on fw fatal recovery flow, we may get\ntx timeout.\n\nFixes: 29429f3300a3 (\"net/mlx5e: Timeout if SQ doesn\u0027t flush during close\")\nSigned-off-by: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\nReviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "9ca415399dae133b00273a4283ef31d003a6818d", "tree": "3cd0a55255ab027973333a58c2a0d80bda92cde2", "parents": [ "6eb7a268a99bad8346d4baa148a14456d061c1c3" ], "author": { "name": "Roi Dayan", "email": "roid@mellanox.com", "time": "Thu May 14 23:44:38 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:52 2020 -0700" }, "message": "net/mlx5: Annotate mutex destroy for root ns\n\nInvoke mutex_destroy() to catch any errors.\n\nFixes: 2cc43b494a6c (\"net/mlx5_core: Managing root flow table\")\nSigned-off-by: Roi Dayan \u003croid@mellanox.com\u003e\nReviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "6eb7a268a99bad8346d4baa148a14456d061c1c3", "tree": "cf0fca9ea39b7a622e40d2871ef9a90333094919", "parents": [ "aee37f3d940ca732df71c3df49347bccaafc0b24" ], "author": { "name": "Roi Dayan", "email": "roid@mellanox.com", "time": "Mon May 11 16:37:11 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:50 2020 -0700" }, "message": "net/mlx5: Don\u0027t maintain a case of del_sw_func being null\n\nAdd del_sw_func cb for root ns. Now there is no need to\nmaintain a case of del_sw_func being null when freeing the node.\n\nFixes: 2cc43b494a6c (\"net/mlx5_core: Managing root flow table\")\nSigned-off-by: Roi Dayan \u003croid@mellanox.com\u003e\nReviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nReviewed-by: Paul Blakey \u003cpaulb@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "aee37f3d940ca732df71c3df49347bccaafc0b24", "tree": "483eed4e1fd43ff7675bb6b32bf049ccecf1dced", "parents": [ "df14ad1eccb04a4a28c90389214dbacab085b244" ], "author": { "name": "Roi Dayan", "email": "roid@mellanox.com", "time": "Mon May 11 16:32:09 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:48 2020 -0700" }, "message": "net/mlx5: Fix cleaning unmanaged flow tables\n\nUnmanaged flow tables doesn\u0027t have a parent and tree_put_node()\nassume there is always a parent if cleaning is needed. fix that.\n\nFixes: 5281a0c90919 (\"net/mlx5: fs_core: Introduce unmanaged flow tables\")\nSigned-off-by: Roi Dayan \u003croid@mellanox.com\u003e\nReviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nReviewed-by: Paul Blakey \u003cpaulb@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "df14ad1eccb04a4a28c90389214dbacab085b244", "tree": "89c396e75f1b1352ff8be3e1217cb3506f176ddc", "parents": [ "a16b8e0dcf7043bee46174bed0553cc9e36b63a5" ], "author": { "name": "Moshe Shemesh", "email": "moshe@mellanox.com", "time": "Wed Apr 29 23:56:58 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:46 2020 -0700" }, "message": "net/mlx5: Fix memory leak in mlx5_events_init\n\nFix memory leak in mlx5_events_init(), in case\ncreate_single_thread_workqueue() fails, events\nstruct should be freed.\n\nFixes: 5d3c537f9070 (\"net/mlx5: Handle event of power detection in the PCIE slot\")\nSigned-off-by: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\nReviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "a16b8e0dcf7043bee46174bed0553cc9e36b63a5", "tree": "e54ff1538791d08cb518567d765a79135569d2e1", "parents": [ "16736e11f43b80a38f98f6add54fab3b8c297df3" ], "author": { "name": "Roi Dayan", "email": "roid@mellanox.com", "time": "Thu Apr 30 09:16:01 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:44 2020 -0700" }, "message": "net/mlx5e: Fix inner tirs handling\n\nIn the cited commit inner_tirs argument was added to create and destroy\ninner tirs, and no indication was added to mlx5e_modify_tirs_hash()\nfunction. In order to have a consistent handling, use\ninner_indir_tir[0].tirn in tirs destroy/modify function as an indication\nto whether inner tirs are created.\nInner tirs are not created for representors and before this commit,\na call to mlx5e_modify_tirs_hash() was sending HW commands to\nmodify non-existent inner tirs.\n\nFixes: 46dc933cee82 (\"net/mlx5e: Provide explicit directive if to create inner indirect tirs\")\nSigned-off-by: Roi Dayan \u003croid@mellanox.com\u003e\nReviewed-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "16736e11f43b80a38f98f6add54fab3b8c297df3", "tree": "4810e55a9368657c9b12bcc3938b36554c0708d6", "parents": [ "321348475d544aa6705dcfac2135deeccb8dc0bb" ], "author": { "name": "Tariq Toukan", "email": "tariqt@mellanox.com", "time": "Mon Apr 27 16:56:59 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:42 2020 -0700" }, "message": "net/mlx5e: kTLS, Destroy key object after destroying the TIS\n\nThe TLS TIS object contains the dek/key ID.\nBy destroying the key first, the TIS would contain an invalid\nnon-existing key ID.\nReverse the destroy order, this also acheives the desired assymetry\nbetween the destroy and the create flows.\n\nFixes: d2ead1f360e8 (\"net/mlx5e: Add kTLS TX HW offload support\")\nSigned-off-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nReviewed-by: Boris Pismenny \u003cborisp@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "321348475d544aa6705dcfac2135deeccb8dc0bb", "tree": "e47bfbd5bbc1a499d2b2b826930d99920f61686e", "parents": [ "f7936ddd35d8b849daf0372770c7c9dbe7910fca" ], "author": { "name": "Maor Dickman", "email": "maord@mellanox.com", "time": "Thu Apr 23 15:16:17 2020 +0300" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:40 2020 -0700" }, "message": "net/mlx5e: Fix allowed tc redirect merged eswitch offload cases\n\nAfter changing the parent_id to be the same for both NICs of same\nThe cited commit wrongly allow offload of tc redirect flows from\nVF to uplink and vice versa when devcies are on different eswitch,\nthese cases aren\u0027t supported by HW.\n\nDisallow the above offloads when devcies are on different eswitch\nand VF LAG is not configured.\n\nFixes: f6dc1264f1c0 (\"net/mlx5e: Disallow tc redirect offload cases we don\u0027t support\")\nSigned-off-by: Maor Dickman \u003cmaord@mellanox.com\u003e\nReviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "f7936ddd35d8b849daf0372770c7c9dbe7910fca", "tree": "2d4d4858da7dcca9d47fef398364747526d39d3c", "parents": [ "d43b7007dbd1195a5b6b83213e49b1516aaf6f5e" ], "author": { "name": "Eran Ben Elisha", "email": "eranbe@mellanox.com", "time": "Thu Mar 19 21:43:13 2020 +0200" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:38 2020 -0700" }, "message": "net/mlx5: Avoid processing commands before cmdif is ready\n\nWhen driver is reloading during recovery flow, it can\u0027t get new commands\ntill command interface is up again. Otherwise we may get to null pointer\ntrying to access non initialized command structures.\n\nAdd cmdif state to avoid processing commands while cmdif is not ready.\n\nFixes: e126ba97dba9 (\"mlx5: Add driver for Mellanox Connect-IB adapters\")\nSigned-off-by: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\nSigned-off-by: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "d43b7007dbd1195a5b6b83213e49b1516aaf6f5e", "tree": "2de29011ba9f66e3883e37253b06e87baaf23a34", "parents": [ "17d00e839d3b592da9659c1977d45f85b77f986a" ], "author": { "name": "Eran Ben Elisha", "email": "eranbe@mellanox.com", "time": "Wed Mar 18 21:44:32 2020 +0200" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:36 2020 -0700" }, "message": "net/mlx5: Fix a race when moving command interface to events mode\n\nAfter driver creates (via FW command) an EQ for commands, the driver will\nbe informed on new commands completion by EQE. However, due to a race in\ndriver\u0027s internal command mode metadata update, some new commands will\nstill be miss-handled by driver as if we are in polling mode. Such commands\ncan get two non forced completion, leading to already freed command entry\naccess.\n\nCREATE_EQ command, that maps EQ to the command queue must be posted to the\ncommand queue while it is empty and no other command should be posted.\n\nAdd SW mechanism that once the CREATE_EQ command is about to be executed,\nall other commands will return error without being sent to the FW. Allow\nsending other commands only after successfully changing the driver\u0027s\ninternal command mode metadata.\nWe can safely return error to all other commands while creating the command\nEQ, as all other commands might be sent from the user/application during\ndriver load. Application can rerun them later after driver\u0027s load was\nfinished.\n\nFixes: e126ba97dba9 (\"mlx5: Add driver for Mellanox Connect-IB adapters\")\nSigned-off-by: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\nSigned-off-by: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "17d00e839d3b592da9659c1977d45f85b77f986a", "tree": "6df648de812dc786f0e89dffe5c4c86198e31fae", "parents": [ "5a730153984dd13f82ffae93d7170d76eba204e9" ], "author": { "name": "Moshe Shemesh", "email": "moshe@mellanox.com", "time": "Fri Dec 27 07:01:53 2019 +0200" }, "committer": { "name": "Saeed Mahameed", "email": "saeedm@mellanox.com", "time": "Fri May 22 17:28:34 2020 -0700" }, "message": "net/mlx5: Add command entry handling completion\n\nWhen FW response to commands is very slow and all command entries in\nuse are waiting for completion we can have a race where commands can get\ntimeout before they get out of the queue and handled. Timeout\ncompletion on uninitialized command will cause releasing command\u0027s\nbuffers before accessing it for initialization and then we will get NULL\npointer exception while trying access it. It may also cause releasing\nbuffers of another command since we may have timeout completion before\neven allocating entry index for this command.\nAdd entry handling completion to avoid this race.\n\nFixes: e126ba97dba9 (\"mlx5: Add driver for Mellanox Connect-IB adapters\")\nSigned-off-by: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\nSigned-off-by: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\nSigned-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n" }, { "commit": "d04322a0da1e86aedaa322ce933cfb8c0191d1eb", "tree": "ef99f5b618065716ff017a5ced9f8402721d0737", "parents": [ "5a730153984dd13f82ffae93d7170d76eba204e9", "f45d01f4f30b53c3a0a1c6c1c154acb7ff74ab9f" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:43:58 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:43:58 2020 -0700" }, "message": "Merge tag \u0027rxrpc-fixes-20200523-v2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs\n\nDavid Howells says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nrxrpc: Fix a warning and a leak [ver #2]\n\nHere are a couple of fixes for AF_RXRPC:\n\n (1) Fix an uninitialised variable warning.\n\n (2) Fix a leak of the ticket on error in rxkad.\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": "f45d01f4f30b53c3a0a1c6c1c154acb7ff74ab9f", "tree": "ef99f5b618065716ff017a5ced9f8402721d0737", "parents": [ "8a1d24e1cc6d96a17f2dcb1400d370cadbfb7cb6" ], "author": { "name": "Qiushi Wu", "email": "wu000273@umn.edu", "time": "Fri May 22 13:45:18 2020 -0500" }, "committer": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Sat May 23 00:35:46 2020 +0100" }, "message": "rxrpc: Fix a memory leak in rxkad_verify_response()\n\nA ticket was not released after a call of the function\n\"rxkad_decrypt_ticket\" failed. Thus replace the jump target\n\"temporary_error_free_resp\" by \"temporary_error_free_ticket\".\n\nFixes: 8c2f826dc3631 (\"rxrpc: Don\u0027t put crypto buffers on the stack\")\nSigned-off-by: Qiushi Wu \u003cwu000273@umn.edu\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\ncc: Markus Elfring \u003cMarkus.Elfring@web.de\u003e\n" }, { "commit": "8a1d24e1cc6d96a17f2dcb1400d370cadbfb7cb6", "tree": "63c6a4716a95791bc86a046f49d8b48095f1abf1", "parents": [ "5a730153984dd13f82ffae93d7170d76eba204e9" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Fri May 22 23:58:28 2020 +0100" }, "committer": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Sat May 23 00:31:39 2020 +0100" }, "message": "rxrpc: Fix a warning\n\nFix a warning due to an uninitialised variable.\n\nle included from ../fs/afs/fs_probe.c:11:\n../fs/afs/fs_probe.c: In function \u0027afs_fileserver_probe_result\u0027:\n../fs/afs/internal.h:1453:2: warning: \u0027rtt_us\u0027 may be used uninitialized in this function [-Wmaybe-uninitialized]\n 1453 | printk(\"[%-6.6s] \"FMT\"\\n\", current-\u003ecomm ,##__VA_ARGS__)\n | ^~~~~~\n../fs/afs/fs_probe.c:35:15: note: \u0027rtt_us\u0027 was declared here\n\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\n" }, { "commit": "5a730153984dd13f82ffae93d7170d76eba204e9", "tree": "42932c9c633d739cfadb6370b84afb2e4cc58ce1", "parents": [ "bf655ba212dfd10d1c86afeee3f3372dbd731d46" ], "author": { "name": "Qiushi Wu", "email": "wu000273@umn.edu", "time": "Fri May 22 16:50:27 2020 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:19:46 2020 -0700" }, "message": "net: sun: fix missing release regions in cas_init_one().\n\nIn cas_init_one(), \"pdev\" is requested by \"pci_request_regions\", but it\nwas not released after a call of the function “pci_write_config_byte”\nfailed. Thus replace the jump target “err_write_cacheline” by\n\"err_out_free_res\".\n\nFixes: 1f26dac32057 (\"[NET]: Add Sun Cassini driver.\")\nSigned-off-by: Qiushi Wu \u003cwu000273@umn.edu\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bf655ba212dfd10d1c86afeee3f3372dbd731d46", "tree": "40c3618974469a5e2dfc7c71d8c98742ba2fa2bd", "parents": [ "561535b0f23961ced071b82575d5e83e6351a814" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Fri May 22 00:31:23 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:15:21 2020 -0700" }, "message": "net: mscc: ocelot: fix address ageing time (again)\n\nocelot_set_ageing_time has 2 callers:\n - felix_set_ageing_time: from drivers/net/dsa/ocelot/felix.c\n - ocelot_port_attr_ageing_set: from drivers/net/ethernet/mscc/ocelot.c\n\nThe issue described in the fixed commit below actually happened for the\nfelix_set_ageing_time code path only, since ocelot_port_attr_ageing_set\nwas already dividing by 1000. So to make both paths symmetrical (and to\nfix addresses getting aged way too fast on Ocelot), stop dividing by\n1000 at caller side altogether.\n\nFixes: c0d7eccbc761 (\"net: mscc: ocelot: ANA_AUTOAGE_AGE_PERIOD holds a value in seconds, not ms\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "561535b0f23961ced071b82575d5e83e6351a814", "tree": "a52c0faf0e485e6bc1958b3cc5ff9ad8d36f24d6", "parents": [ "156ee62bbd8d3234aef26c68f4ba022217591149" ], "author": { "name": "Heiner Kallweit", "email": "hkallweit1@gmail.com", "time": "Thu May 21 22:03:08 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:14:01 2020 -0700" }, "message": "r8169: fix OCP access on RTL8117\n\nAccording to r8168 vendor driver DASHv3 chips like RTL8168fp/RTL8117\nneed a special addressing for OCP access.\nFix is compile-tested only due to missing test hardware.\n\nFixes: 1287723aa139 (\"r8169: add support for RTL8117\")\nSigned-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "156ee62bbd8d3234aef26c68f4ba022217591149", "tree": "5c11a9e8bc90a1f20433d186f043aacb17559e28", "parents": [ "a96ac8a0045e3cbe3e5af6d1b3c78c6c2065dec5", "46ca11177ed593f39d534f8d2c74ec5344e90c11" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:08:14 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:08:14 2020 -0700" }, "message": "Merge branch \u0027mlxsw-Various-fixes\u0027\n\nIdo Schimmel says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nmlxsw: Various fixes\n\nPatch #1 from Jiri fixes a use-after-free discovered while fuzzing mlxsw\n/ devlink with syzkaller.\n\nPatch #2 from Amit works around a limitation in new versions of arping,\nwhich is used in several selftests.\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": "46ca11177ed593f39d534f8d2c74ec5344e90c11", "tree": "5c11a9e8bc90a1f20433d186f043aacb17559e28", "parents": [ "4340f42f207eacb81e7a6b6bb1e3b6afad9a2e26" ], "author": { "name": "Amit Cohen", "email": "amitc@mellanox.com", "time": "Thu May 21 15:11:45 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:08:14 2020 -0700" }, "message": "selftests: mlxsw: qos_mc_aware: Specify arping timeout as an integer\n\nStarting from iputils s20190709 (used in Fedora 31), arping does not\nsupport timeout being specified as a decimal:\n\n$ arping -c 1 -I swp1 -b 192.0.2.66 -q -w 0.1\narping: invalid argument: \u00270.1\u0027\n\nPreviously, such timeouts were rounded to an integer.\n\nFix this by specifying the timeout as an integer.\n\nFixes: a5ee171d087e (\"selftests: mlxsw: qos_mc_aware: Add a test for UC awareness\")\nSigned-off-by: Amit Cohen \u003camitc@mellanox.com\u003e\nReviewed-by: Petr Machata \u003cpetrm@mellanox.com\u003e\nSigned-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4340f42f207eacb81e7a6b6bb1e3b6afad9a2e26", "tree": "605361b0e9f273e1eea08b80644a631665c27c34", "parents": [ "a96ac8a0045e3cbe3e5af6d1b3c78c6c2065dec5" ], "author": { "name": "Jiri Pirko", "email": "jiri@mellanox.com", "time": "Thu May 21 15:11:44 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:08:14 2020 -0700" }, "message": "mlxsw: spectrum: Fix use-after-free of split/unsplit/type_set in case reload fails\n\nIn case of reload fail, the mlxsw_sp-\u003eports contains a pointer to a\nfreed memory (either by reload_down() or reload_up() error path).\nFix this by initializing the pointer to NULL and checking it before\ndereferencing in split/unsplit/type_set callpaths.\n\nFixes: 24cc68ad6c46 (\"mlxsw: core: Add support for reload\")\nReported-by: Danielle Ratson \u003cdanieller@mellanox.com\u003e\nSigned-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a96ac8a0045e3cbe3e5af6d1b3c78c6c2065dec5", "tree": "6d13f31bd478131d4f6e28ba9f686c5bca2fdb50", "parents": [ "7a40a2d22b79131a4826777864f39b19eaf7c857" ], "author": { "name": "Jonathan McDowell", "email": "noodles@earth.li", "time": "Thu May 21 12:49:34 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:07:02 2020 -0700" }, "message": "net: ethernet: stmmac: Enable interface clocks on probe for IPQ806x\n\nThe ipq806x_gmac_probe() function enables the PTP clock but not the\nappropriate interface clocks. This means that if the bootloader hasn\u0027t\ndone so attempting to bring up the interface will fail with an error\nlike:\n\n[ 59.028131] ipq806x-gmac-dwmac 37600000.ethernet: Failed to reset the dma\n[ 59.028196] ipq806x-gmac-dwmac 37600000.ethernet eth1: stmmac_hw_setup: DMA engine initialization failed\n[ 59.034056] ipq806x-gmac-dwmac 37600000.ethernet eth1: stmmac_open: Hw setup failed\n\nThis patch, a slightly cleaned up version of one posted by Sergey\nSergeev in:\n\nhttps://forum.openwrt.org/t/support-for-mikrotik-rb3011uias-rm/4064/257\n\ncorrectly enables the clock; we have already configured the source just\nbefore this.\n\nTested on a MikroTik RB3011.\n\nSigned-off-by: Jonathan McDowell \u003cnoodles@earth.li\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7a40a2d22b79131a4826777864f39b19eaf7c857", "tree": "53a3c68f7272cfb870d352a6f5b8897b528d90e9", "parents": [ "4629ed2e481b2af421ccb2ca3ba6e6b541789f5a", "4d59e59cf45046fe1263a935f8abc418bb61215c" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:05:42 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:05:42 2020 -0700" }, "message": "Merge branch \u0027netdevsim-Two-small-fixes\u0027\n\nIdo Schimmel says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnetdevsim: Two small fixes\n\nFix two bugs observed while analyzing regression failures.\n\nPatch #1 fixes a bug where sometimes the drop counter of a packet trap\npolicer would not increase.\n\nPatch #2 adds a missing initialization of a variable in a related\nselftest.\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": "4d59e59cf45046fe1263a935f8abc418bb61215c", "tree": "53a3c68f7272cfb870d352a6f5b8897b528d90e9", "parents": [ "be43224fc0e4697ad0d03107cbaf1ecf26e7ee72" ], "author": { "name": "Ido Schimmel", "email": "idosch@mellanox.com", "time": "Thu May 21 14:46:17 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:05:42 2020 -0700" }, "message": "selftests: netdevsim: Always initialize \u0027RET\u0027 variable\n\nThe variable is used by log_test() to check if the test case completely\nsuccessfully or not. In case it is not initialized at the start of a\ntest case, it is possible for the test case to fail despite not\nencountering any errors.\n\nExample:\n\n```\n...\nTEST: Trap group statistics [ OK ]\nTEST: Trap policer [FAIL]\n\tPolicer drop counter was not incremented\nTEST: Trap policer binding [FAIL]\n\tPolicer drop counter was not incremented\n```\n\nFailure of trap_policer_test() caused trap_policer_bind_test() to fail\nas well.\n\nFix by adding missing initialization of the variable.\n\nFixes: 5fbff58e27a1 (\"selftests: netdevsim: Add test cases for devlink-trap policers\")\nSigned-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\nReviewed-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "be43224fc0e4697ad0d03107cbaf1ecf26e7ee72", "tree": "84a142dc5f23911f90d51d26d30717d46391aeae", "parents": [ "4629ed2e481b2af421ccb2ca3ba6e6b541789f5a" ], "author": { "name": "Ido Schimmel", "email": "idosch@mellanox.com", "time": "Thu May 21 14:46:16 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 16:05:42 2020 -0700" }, "message": "netdevsim: Ensure policer drop counter always increases\n\nIn case the policer drop counter is retrieved when the jiffies value is\na multiple of 64, the counter will not be incremented.\n\nThis randomly breaks a selftest [1] the reads the counter twice and\nchecks that it was incremented:\n\n```\nTEST: Trap policer [FAIL]\n\tPolicer drop counter was not incremented\n```\n\nFix by always incrementing the counter by 1.\n\n[1] tools/testing/selftests/drivers/net/netdevsim/devlink_trap.sh\n\nFixes: ad188458d012 (\"netdevsim: Add devlink-trap policer support\")\nSigned-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\nReviewed-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4629ed2e481b2af421ccb2ca3ba6e6b541789f5a", "tree": "56e58d926d4553ccb781eb270287ae6f200137e0", "parents": [ "79dde73cf9bcf1dd317a2667f78b758e9fe139ed", "441fdee1eaf050ef0040bde0d7af075c1c6a6d8b" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 15:53:30 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 15:53:30 2020 -0700" }, "message": "Merge tag \u0027rxrpc-fixes-20200520\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs\n\nDavid Howells says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nrxrpc: Fix retransmission timeout and ACK discard\n\nHere are a couple of fixes and an extra tracepoint for AF_RXRPC:\n\n (1) Calculate the RTO pretty much as TCP does, rather than making\n something up, including an initial 4s timeout (which causes return\n probes from the fileserver to fail if a packet goes missing), and add\n backoff.\n\n (2) Fix the discarding of out-of-order received ACKs. We mustn\u0027t let the\n hard-ACK point regress, nor do we want to do unnecessary\n retransmission because the soft-ACK list regresses. This is not\n trivial, however, due to some loose wording in various old protocol\n specs, the ACK field that should be used for this sometimes has the\n wrong information in it.\n\n (3) Add a tracepoint to log a discarded ACK.\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": "79dde73cf9bcf1dd317a2667f78b758e9fe139ed", "tree": "293dd8aac7ff1ad1d16cb314825389e0ebd8acda", "parents": [ "d3e8e4c11870413789f029a71e72ae6e971fe678" ], "author": { "name": "Valentin Longchamp", "email": "valentin@longchamp.me", "time": "Wed May 20 17:53:50 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 15:50:34 2020 -0700" }, "message": "net/ethernet/freescale: rework quiesce/activate for ucc_geth\n\nugeth_quiesce/activate are used to halt the controller when there is a\nlink change that requires to reconfigure the mac.\n\nThe previous implementation called netif_device_detach(). This however\ncauses the initial activation of the netdevice to fail precisely because\nit\u0027s detached. For details, see [1].\n\nA possible workaround was the revert of commit\nnet: linkwatch: add check for netdevice being present to linkwatch_do_dev\nHowever, the check introduced in the above commit is correct and shall be\nkept.\n\nThe netif_device_detach() is thus replaced with\nnetif_tx_stop_all_queues() that prevents any tranmission. This allows to\nperform mac config change required by the link change, without detaching\nthe corresponding netdevice and thus not preventing its initial\nactivation.\n\n[1] https://lists.openwall.net/netdev/2020/01/08/201\n\nSigned-off-by: Valentin Longchamp \u003cvalentin@longchamp.me\u003e\nAcked-by: Matteo Ghidoni \u003cmatteo.ghidoni@ch.abb.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d3e8e4c11870413789f029a71e72ae6e971fe678", "tree": "0f42e8454d44e931a9018ff1c5108dad59e4bbe4", "parents": [ "1378817486d6860f6a927f573491afe65287abf1" ], "author": { "name": "Jere Leppänen", "email": "jere.leppanen@nokia.com", "time": "Wed May 20 18:15:31 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 15:47:59 2020 -0700" }, "message": "sctp: Start shutdown on association restart if in SHUTDOWN-SENT state and socket is closed\n\nCommit bdf6fa52f01b (\"sctp: handle association restarts when the\nsocket is closed.\") starts shutdown when an association is restarted,\nif in SHUTDOWN-PENDING state and the socket is closed. However, the\nrationale stated in that commit applies also when in SHUTDOWN-SENT\nstate - we don\u0027t want to move an association to ESTABLISHED state when\nthe socket has been closed, because that results in an association\nthat is unreachable from user space.\n\nThe problem scenario:\n\n1. Client crashes and/or restarts.\n\n2. Server (using one-to-one socket) calls close(). SHUTDOWN is lost.\n\n3. Client reconnects using the same addresses and ports.\n\n4. Server\u0027s association is restarted. The association and the socket\n move to ESTABLISHED state, even though the server process has\n closed its descriptor.\n\nAlso, after step 4 when the server process exits, some resources are\nleaked in an attempt to release the underlying inet sock structure in\nESTABLISHED state:\n\n IPv4: Attempt to release TCP socket in state 1 00000000377288c7\n\nFix by acting the same way as in SHUTDOWN-PENDING state. That is, if\nan association is restarted in SHUTDOWN-SENT state and the socket is\nclosed, then start shutdown and don\u0027t move the association or the\nsocket to ESTABLISHED state.\n\nFixes: bdf6fa52f01b (\"sctp: handle association restarts when the socket is closed.\")\nSigned-off-by: Jere Leppänen \u003cjere.leppanen@nokia.com\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1378817486d6860f6a927f573491afe65287abf1", "tree": "c3b3204db48706dc935dd13bd08705c4cab9c085", "parents": [ "3138a07ce219acde4c0d7ea0b6d54ba64153328b" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu May 21 11:29:58 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 15:39:00 2020 -0700" }, "message": "tipc: block BH before using dst_cache\n\ndst_cache_get() documents it must be used with BH disabled.\n\nsysbot reported :\n\nBUG: using smp_processor_id() in preemptible [00000000] code: /21697\ncaller is dst_cache_get+0x3a/0xb0 net/core/dst_cache.c:68\nCPU: 0 PID: 21697 Comm: Not tainted 5.7.0-rc6-syzkaller #0\nHardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011\nCall Trace:\n __dump_stack lib/dump_stack.c:77 [inline]\n dump_stack+0x188/0x20d lib/dump_stack.c:118\n check_preemption_disabled lib/smp_processor_id.c:47 [inline]\n debug_smp_processor_id.cold+0x88/0x9b lib/smp_processor_id.c:57\n dst_cache_get+0x3a/0xb0 net/core/dst_cache.c:68\n tipc_udp_xmit.isra.0+0xb9/0xad0 net/tipc/udp_media.c:164\n tipc_udp_send_msg+0x3e6/0x490 net/tipc/udp_media.c:244\n tipc_bearer_xmit_skb+0x1de/0x3f0 net/tipc/bearer.c:526\n tipc_enable_bearer+0xb2f/0xd60 net/tipc/bearer.c:331\n __tipc_nl_bearer_enable+0x2bf/0x390 net/tipc/bearer.c:995\n tipc_nl_bearer_enable+0x1e/0x30 net/tipc/bearer.c:1003\n genl_family_rcv_msg_doit net/netlink/genetlink.c:673 [inline]\n genl_family_rcv_msg net/netlink/genetlink.c:718 [inline]\n genl_rcv_msg+0x627/0xdf0 net/netlink/genetlink.c:735\n netlink_rcv_skb+0x15a/0x410 net/netlink/af_netlink.c:2469\n genl_rcv+0x24/0x40 net/netlink/genetlink.c:746\n netlink_unicast_kernel net/netlink/af_netlink.c:1303 [inline]\n netlink_unicast+0x537/0x740 net/netlink/af_netlink.c:1329\n netlink_sendmsg+0x882/0xe10 net/netlink/af_netlink.c:1918\n sock_sendmsg_nosec net/socket.c:652 [inline]\n sock_sendmsg+0xcf/0x120 net/socket.c:672\n ____sys_sendmsg+0x6bf/0x7e0 net/socket.c:2362\n ___sys_sendmsg+0x100/0x170 net/socket.c:2416\n __sys_sendmsg+0xec/0x1b0 net/socket.c:2449\n do_syscall_64+0xf6/0x7d0 arch/x86/entry/common.c:295\n entry_SYSCALL_64_after_hwframe+0x49/0xb3\nRIP: 0033:0x45ca29\n\nFixes: e9c1a793210f (\"tipc: add dst_cache support for udp media\")\nCc: Xin Long \u003clucien.xin@gmail.com\u003e\nCc: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReported-by: syzbot \u003csyzkaller@googlegroups.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3138a07ce219acde4c0d7ea0b6d54ba64153328b", "tree": "04fb7f34a251288b18c8eb772ff0b1df899a9976", "parents": [ "d3b968bc2dee06a88682d035a0fde8bf77af39b9" ], "author": { "name": "Russell King", "email": "rmk+kernel@armlinux.org.uk", "time": "Wed May 20 12:26:35 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 15:31:43 2020 -0700" }, "message": "net: mvpp2: fix RX hashing for non-10G ports\n\nWhen rxhash is enabled on any ethernet port except the first in each CP\nblock, traffic flow is prevented. The analysis is below:\n\nI\u0027ve been investigating this afternoon, and what I\u0027ve found, comparing\na kernel without 895586d5dc32 and with 895586d5dc32 applied is:\n\n- The table programmed into the hardware via mvpp22_rss_fill_table()\n appears to be identical with or without the commit.\n\n- When rxhash is enabled on eth2, mvpp2_rss_port_c2_enable() reports\n that c2.attr[0] and c2.attr[2] are written back containing:\n\n - with 895586d5dc32, failing: 00200000 40000000\n - without 895586d5dc32, working: 04000000 40000000\n\n- When disabling rxhash, c2.attr[0] and c2.attr[2] are written back as:\n\n 04000000 00000000\n\nThe second value represents the MVPP22_CLS_C2_ATTR2_RSS_EN bit, the\nfirst value is the queue number, which comprises two fields. The high\n5 bits are 24:29 and the low three are 21:23 inclusive. This comes\nfrom:\n\n c2.attr[0] \u003d MVPP22_CLS_C2_ATTR0_QHIGH(qh) |\n MVPP22_CLS_C2_ATTR0_QLOW(ql);\n\nSo, the working case gives eth2 a queue id of 4.0, or 32 as per\nport-\u003efirst_rxq, and the non-working case a queue id of 0.1, or 1.\nThe allocation of queue IDs seems to be in mvpp2_port_probe():\n\n if (priv-\u003ehw_version \u003d\u003d MVPP21)\n port-\u003efirst_rxq \u003d port-\u003eid * port-\u003enrxqs;\n else\n port-\u003efirst_rxq \u003d port-\u003eid * priv-\u003emax_port_rxqs;\n\nWhere:\n\n if (priv-\u003ehw_version \u003d\u003d MVPP21)\n priv-\u003emax_port_rxqs \u003d 8;\n else\n priv-\u003emax_port_rxqs \u003d 32;\n\nMaking the port 0 (eth0 / eth1) have port-\u003efirst_rxq \u003d 0, and port 1\n(eth2) be 32. It seems the idea is that the first 32 queues belong to\nport 0, the second 32 queues belong to port 1, etc.\n\nmvpp2_rss_port_c2_enable() gets the queue number from it\u0027s parameter,\n\u0027ctx\u0027, which comes from mvpp22_rss_ctx(port, 0). This returns\nport-\u003erss_ctx[0].\n\nmvpp22_rss_context_create() is responsible for allocating that, which\nit does by looking for an unallocated priv-\u003erss_tables[] pointer. This\ntable is shared amongst all ports on the CP silicon.\n\nWhen we write the tables in mvpp22_rss_fill_table(), the RSS table\nentry is defined by:\n\n u32 sel \u003d MVPP22_RSS_INDEX_TABLE(rss_ctx) |\n MVPP22_RSS_INDEX_TABLE_ENTRY(i);\n\nwhere rss_ctx is the context ID (queue number) and i is the index in\nthe table.\n\nIf we look at what is written:\n\n- The first table to be written has \"sel\" values of 00000000..0000001f,\n containing values 0..3. This appears to be for eth1. This is table 0,\n RX queue number 0.\n- The second table has \"sel\" values of 00000100..0000011f, and appears\n to be for eth2. These contain values 0x20..0x23. This is table 1,\n RX queue number 0.\n- The third table has \"sel\" values of 00000200..0000021f, and appears\n to be for eth3. These contain values 0x40..0x43. This is table 2,\n RX queue number 0.\n\nHow do queue numbers translate to the RSS table? There is another\ntable - the RXQ2RSS table, indexed by the MVPP22_RSS_INDEX_QUEUE field\nof MVPP22_RSS_INDEX and accessed through the MVPP22_RXQ2RSS_TABLE\nregister. Before 895586d5dc32, it was:\n\n mvpp2_write(priv, MVPP22_RSS_INDEX,\n MVPP22_RSS_INDEX_QUEUE(port-\u003efirst_rxq));\n mvpp2_write(priv, MVPP22_RXQ2RSS_TABLE,\n MVPP22_RSS_TABLE_POINTER(port-\u003eid));\n\nand after:\n\n mvpp2_write(priv, MVPP22_RSS_INDEX, MVPP22_RSS_INDEX_QUEUE(ctx));\n mvpp2_write(priv, MVPP22_RXQ2RSS_TABLE, MVPP22_RSS_TABLE_POINTER(ctx));\n\nBefore the commit, for eth2, that would\u0027ve contained \u002732\u0027 for the\nindex and \u00271\u0027 for the table pointer - mapping queue 32 to table 1.\nRemember that this is queue-high.queue-low of 4.0.\n\nAfter the commit, we appear to map queue 1 to table 1. That again\nlooks fine on the face of it.\n\nSection 9.3.1 of the A8040 manual seems indicate the reason that the\nqueue number is separated. queue-low seems to always come from the\nclassifier, whereas queue-high can be from the ingress physical port\nnumber or the classifier depending on the MVPP2_CLS_SWFWD_PCTRL_REG.\n\nWe set the port bit in MVPP2_CLS_SWFWD_PCTRL_REG, meaning that queue-high\ncomes from the MVPP2_CLS_SWFWD_P2HQ_REG() register... and this seems to\nbe where our bug comes from.\n\nmvpp2_cls_oversize_rxq_set() sets this up as:\n\n mvpp2_write(port-\u003epriv, MVPP2_CLS_SWFWD_P2HQ_REG(port-\u003eid),\n (port-\u003efirst_rxq \u003e\u003e MVPP2_CLS_OVERSIZE_RXQ_LOW_BITS));\n\n val \u003d mvpp2_read(port-\u003epriv, MVPP2_CLS_SWFWD_PCTRL_REG);\n val |\u003d MVPP2_CLS_SWFWD_PCTRL_MASK(port-\u003eid);\n mvpp2_write(port-\u003epriv, MVPP2_CLS_SWFWD_PCTRL_REG, val);\n\nSetting the MVPP2_CLS_SWFWD_PCTRL_MASK bit means that the queue-high\nfor eth2 is _always_ 4, so only queues 32 through 39 inclusive are\navailable to eth2. Yet, we\u0027re trying to tell the classifier to set\nqueue-high, which will be ignored, to zero. Hence, the queue-high\nfield (MVPP22_CLS_C2_ATTR0_QHIGH()) from the classifier will be\nignored.\n\nThis means we end up directing traffic from eth2 not to queue 1, but\nto queue 33, and then we tell it to look up queue 33 in the RSS table.\nHowever, RSS table has not been programmed for queue 33, and so it ends\nup (presumably) dropping the packets.\n\nIt seems that mvpp22_rss_context_create() doesn\u0027t take account of the\nfact that the upper 5 bits of the queue ID can\u0027t actually be changed\ndue to the settings in mvpp2_cls_oversize_rxq_set(), _or_ it seems that\nmvpp2_cls_oversize_rxq_set() has been missed in this commit. Either\nway, these two functions mutually disagree with what queue number\nshould be used.\n\nLooking deeper into what mvpp2_cls_oversize_rxq_set() and the MTU\nvalidation is doing, it seems that MVPP2_CLS_SWFWD_P2HQ_REG() is used\nfor over-sized packets attempting to egress through this port. With\nthe classifier having had RSS enabled and directing eth2 traffic to\nqueue 1, we may still have packets appearing on queue 32 for this port.\n\nHowever, the only way we may end up with over-sized packets attempting\nto egress through eth2 - is if the A8040 forwards frames between its\nports. From what I can see, we don\u0027t support that feature, and the\nkernel restricts the egress packet size to the MTU. In any case, if we\nwere to attempt to transmit an oversized packet, we have no support in\nthe kernel to deal with that appearing in the port\u0027s receive queue.\n\nSo, this patch attempts to solve the issue by clearing the\nMVPP2_CLS_SWFWD_PCTRL_MASK() bit, allowing MVPP22_CLS_C2_ATTR0_QHIGH()\nfrom the classifier to define the queue-high field of the queue number.\n\nMy testing seems to confirm my findings above - clearing this bit\nmeans that if I enable rxhash on eth2, the interface can then pass\ntraffic, as we are now directing traffic to RX queue 1 rather than\nqueue 33. Traffic still seems to work with rxhash off as well.\n\nReported-by: Matteo Croce \u003cmcroce@redhat.com\u003e\nTested-by: Matteo Croce \u003cmcroce@redhat.com\u003e\nFixes: 895586d5dc32 (\"net: mvpp2: cls: Use RSS contexts to handle RSS tables\")\nSigned-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d3b968bc2dee06a88682d035a0fde8bf77af39b9", "tree": "2cd5e308629564bb49756516568f819781f69539", "parents": [ "b4024c9e5c57902155d3b5e7de482e245f492bff", "5cf65922bb15279402e1e19b5ee8c51d618fa51f" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 14:35:35 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 14:35:35 2020 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf\n\nDaniel Borkmann says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\npull-request: bpf 2020-05-22\n\nThe following pull-request contains BPF updates for your *net* tree.\n\nWe\u0027ve added 3 non-merge commits during the last 3 day(s) which contain\na total of 5 files changed, 69 insertions(+), 11 deletions(-).\n\nThe main changes are:\n\n1) Fix to reject mmap()\u0027ing read-only array maps as writable since BPF verifier\n relies on such map content to be frozen, from Andrii Nakryiko.\n\n2) Fix breaking audit from secid_to_secctx() LSM hook by avoiding to use\n call_int_hook() since this hook is not stackable, from KP Singh.\n\n3) Fix BPF flow dissector program ref leak on netns cleanup, from Jakub Sitnicki.\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": "b4024c9e5c57902155d3b5e7de482e245f492bff", "tree": "c3141855fb97a4fece2d48ae6dcbcd1e6a39aeae", "parents": [ "bd6972226f50910a5b97e6b9d443c5d0433bf054" ], "author": { "name": "Claudiu Manoil", "email": "claudiu.manoil@nxp.com", "time": "Fri May 22 11:54:34 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 14:24:46 2020 -0700" }, "message": "felix: Fix initialization of ioremap resources\n\nThe caller of devm_ioremap_resource(), either accidentally\nor by wrong assumption, is writing back derived resource data\nto global static resource initialization tables that should\nhave been constant. Meaning that after it computes the final\nphysical start address it saves the address for no reason\nin the static tables. This doesn\u0027t affect the first driver\nprobing after reboot, but it breaks consecutive driver reloads\n(i.e. driver unbind \u0026 bind) because the initialization tables\nno longer have the correct initial values. So the next probe()\nwill map the device registers to wrong physical addresses,\ncausing ARM SError async exceptions.\nThis patch fixes all of the above.\n\nFixes: 56051948773e (\"net: dsa: ocelot: add driver for Felix switch family\")\nSigned-off-by: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\nReviewed-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nTested-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bd6972226f50910a5b97e6b9d443c5d0433bf054", "tree": "85bd706c6bdb3bc09e76fd86271b955dd0b08743", "parents": [ "a7654211d0ffeaa8eb0545ea00f8445242cbce05" ], "author": { "name": "Todd Malsbary", "email": "todd.malsbary@linux.intel.com", "time": "Thu May 21 19:10:49 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 22 14:21:24 2020 -0700" }, "message": "mptcp: use untruncated hash in ADD_ADDR HMAC\n\nThere is some ambiguity in the RFC as to whether the ADD_ADDR HMAC is\nthe rightmost 64 bits of the entire hash or of the leftmost 160 bits\nof the hash. The intention, as clarified with the author of the RFC,\nis the entire hash.\n\nThis change returns the entire hash from\nmptcp_crypto_hmac_sha (instead of only the first 160 bits), and moves\nany truncation/selection operation on the hash to the caller.\n\nFixes: 12555a2d97e5 (\"mptcp: use rightmost 64 bits in ADD_ADDR HMAC\")\nReviewed-by: Christoph Paasch \u003ccpaasch@apple.com\u003e\nReviewed-by: Mat Martineau \u003cmathew.j.martineau@linux.intel.com\u003e\nSigned-off-by: Todd Malsbary \u003ctodd.malsbary@linux.intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "444565650a5fe9c63ddf153e6198e31705dedeb2", "tree": "8b63ac63f509d4afef2ac7554d519a77aa5cde83", "parents": [ "db9f384785b15e8569910c8e2ea1ffc83aedbff6", "d4ae271dfaae2a5f41c015f2f20d62a1deeec734" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 11:12:30 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 11:12:30 2020 -0700" }, "message": "Merge tag \u0027io_uring-5.7-2020-05-22\u0027 of git://git.kernel.dk/linux-block\n\nPull io_uring fixes from Jens Axboe:\n \"A small collection of small fixes that should go into this release:\n\n - Two fixes for async request preparation (Pavel)\n\n - Busy clear fix for SQPOLL (Xiaoguang)\n\n - Don\u0027t use kiocb-\u003eprivate for O_DIRECT buf index, some file systems\n use it (Bijan)\n\n - Kill dead check in io_splice()\n\n - Ensure sqo_wait is initialized early\n\n - Cancel task_work if we fail adding to original process\n\n - Only add (IO)pollable requests to iopoll list, fixing a regression\n in this merge window\"\n\n* tag \u0027io_uring-5.7-2020-05-22\u0027 of git://git.kernel.dk/linux-block:\n io_uring: reset -EBUSY error when io sq thread is waken up\n io_uring: don\u0027t add non-IO requests to iopoll pending list\n io_uring: don\u0027t use kiocb.private to store buf_index\n io_uring: cancel work if task_work_add() fails\n io_uring: remove dead check in io_splice()\n io_uring: fix FORCE_ASYNC req preparation\n io_uring: don\u0027t prepare DRAIN reqs twice\n io_uring: initialize ctx-\u003esqo_wait earlier\n" }, { "commit": "db9f384785b15e8569910c8e2ea1ffc83aedbff6", "tree": "069751bd1ade93cdcb8ad0aaae2bbc1fa95212e2", "parents": [ "b09ca17a2a082299c787025b6c3e614a24a24ca8", "1592cd15eec6e2952453f9a82da6e8a53e2b8db5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 11:10:42 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 11:10:42 2020 -0700" }, "message": "Merge tag \u0027block-5.7-2020-05-22\u0027 of git://git.kernel.dk/linux-block\n\nPull block fixes from Jens Axboe:\n \"Two fixes for null_blk zone mode\"\n\n* tag \u0027block-5.7-2020-05-22\u0027 of git://git.kernel.dk/linux-block:\n null_blk: don\u0027t allow discard for zoned mode\n null_blk: return error for invalid zone size\n" }, { "commit": "b09ca17a2a082299c787025b6c3e614a24a24ca8", "tree": "1aaf05ea769347691996c8626240ef170245b674", "parents": [ "4286d192c803571e8ca43b0f1f8ea04d663a278a", "8356c379cfba8b1b90b0a2423f6afbbe2cdc5d91" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 11:08:26 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 11:08:26 2020 -0700" }, "message": "Merge tag \u0027riscv-for-linus-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux\n\nPull RISC-V fixes from Palmer Dabbelt:\n \"Two fixes:\n\n - Another !MMU build fix that was a straggler from last week\n\n - A fix to use the \"register\" keyword for the GP global register\n variable\"\n\n* tag \u0027riscv-for-linus-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:\n RISC-V: gp_in_global needs register keyword\n riscv: Fix print_vm_layout build error if NOMMU\n" }, { "commit": "9bb4cbf4862dfa139f73e12912129e5b76baea1a", "tree": "281445c282163ff286d7eb5b6b60f1a5eb608625", "parents": [ "b9bbe6ed63b2b9f2c9ee5cbd0f2c946a2723f4ce", "b4f1874c62168159fdb419ced4afc77c1b51c475" ], "author": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Fri May 22 20:06:25 2020 +0200" }, "committer": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Fri May 22 20:06:25 2020 +0200" }, "message": "Merge tag \u0027efi-fixes-for-v5.7-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/efi/efi into efi/urgent\n\nPull EFI fixes from Ard Biesheuvel:\n\n\"- fix EFI framebuffer earlycon for wide fonts\n - avoid filling screen_info with garbage if the EFI framebuffer is not\n available\n - fix a potential host tool build error due to a symbol clash on x86\n - work around a EFI firmware bug regarding the binary format of the TPM\n final events table\n - fix a missing memory free by reworking the E820 table sizing routine to\n not do the allocation in the first place\n - add CPER parsing for firmware errors\"\n" }, { "commit": "187b96db5ca79423618dfa29a05c438c34f9e1f0", "tree": "f5eea6c1e32249cfab3bc512fbe136e85d691bcc", "parents": [ "d7110a26e5905ec2fe3fc88bc6a538901accb72b" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@redhat.com", "time": "Fri May 22 08:54:35 2020 -0500" }, "committer": { "name": "Peter Zijlstra", "email": "peterz@infradead.org", "time": "Fri May 22 19:55:17 2020 +0200" }, "message": "x86/unwind/orc: Fix unwind_get_return_address_ptr() for inactive tasks\n\nNormally, show_trace_log_lvl() scans the stack, looking for text\naddresses to print. In parallel, it unwinds the stack with\nunwind_next_frame(). If the stack address matches the pointer returned\nby unwind_get_return_address_ptr() for the current frame, the text\naddress is printed normally without a question mark. Otherwise it\u0027s\nconsidered a breadcrumb (potentially from a previous call path) and it\u0027s\nprinted with a question mark to indicate that the address is unreliable\nand typically can be ignored.\n\nSince the following commit:\n\n f1d9a2abff66 (\"x86/unwind/orc: Don\u0027t skip the first frame for inactive tasks\")\n\n... for inactive tasks, show_trace_log_lvl() prints *only* unreliable\naddresses (prepended with \u0027?\u0027).\n\nThat happens because, for the first frame of an inactive task,\nunwind_get_return_address_ptr() returns the wrong return address\npointer: one word *below* the task stack pointer. show_trace_log_lvl()\nstarts scanning at the stack pointer itself, so it never finds the first\n\u0027reliable\u0027 address, causing only guesses to being printed.\n\nThe first frame of an inactive task isn\u0027t a normal stack frame. It\u0027s\nactually just an instance of \u0027struct inactive_task_frame\u0027 which is left\nbehind by __switch_to_asm(). Now that this inactive frame is actually\nexposed to callers, fix unwind_get_return_address_ptr() to interpret it\nproperly.\n\nFixes: f1d9a2abff66 (\"x86/unwind/orc: Don\u0027t skip the first frame for inactive tasks\")\nReported-by: Tetsuo Handa \u003cpenguin-kernel@i-love.sakura.ne.jp\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nSigned-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nLink: https://lkml.kernel.org/r/20200522135435.vbxs7umku5pyrdbk@treble\n" }, { "commit": "4286d192c803571e8ca43b0f1f8ea04d663a278a", "tree": "e3ec5d9cfddfa6465c27a046cc0a01ddccae80ac", "parents": [ "f5ca7a7161028db5ac5bcd06db03d7f2e340a20d", "8cfb347ad0cffdbfc69c82506fb3be9429563211" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 09:34:19 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 09:34:19 2020 -0700" }, "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 - Bring the PTRACE_SYSEMU semantics in line with the man page.\n\n - Annotate variable assignment in get_user() with the type to avoid\n sparse warnings.\n\n* tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:\n arm64: Add get_user() type annotation on the !access_ok() path\n arm64: Fix PTRACE_SYSEMU semantics\n" }, { "commit": "f5ca7a7161028db5ac5bcd06db03d7f2e340a20d", "tree": "8059c861291aceeaa6f4c585d14fc9757e94c135", "parents": [ "c8347bbf19f265c1bd254ca148f27caa71e77d61", "259eb82475316672a5d682a94dc8bdd53cf8d8c3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 09:22:22 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 09:22:22 2020 -0700" }, "message": "Merge tag \u0027sound-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound\n\nPull sound fixes from Takashi Iwai:\n \"Just a few small fixes: the only significant one is a slight\n improvement for PCM running position update with no-period-elapsed\n case while the rest are HD-audio fixups and ice1712 model quirk\"\n\n* tag \u0027sound-5.7-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:\n ALSA: hda/realtek - Add more fixup entries for Clevo machines\n ALSA: iec1712: Initialize STDSP24 properly when using the model\u003dstaudio option\n ALSA: hda/realtek - Fix silent output on Gigabyte X570 Aorus Xtreme\n ALSA: pcm: fix incorrect hw_base increase\n" }, { "commit": "8cfb347ad0cffdbfc69c82506fb3be9429563211", "tree": "68572474dac4176aab8a86a63b11a03a8f57e005", "parents": [ "1cf6022bd9161081215028203919c33fcfa6debb" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Fri May 22 15:23:21 2020 +0100" }, "committer": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Fri May 22 16:59:49 2020 +0100" }, "message": "arm64: Add get_user() type annotation on the !access_ok() path\n\nSparse reports \"Using plain integer as NULL pointer\" when the arm64\n__get_user_error() assigns 0 to a pointer type. Use proper type\nannotation.\n\nSigned-of-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nReported-by: kbuild test robot \u003clkp@intel.com\u003e\nLink: http://lkml.kernel.org/r/20200522142321.GP23230@ZenIV.linux.org.uk\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n" }, { "commit": "c8347bbf19f265c1bd254ca148f27caa71e77d61", "tree": "06b20776bda5f6f2e5b801ebd010dbd028332928", "parents": [ "051143e1602d90ea71887d92363edd539d411de5", "8659a0e0efdd975c73355dbc033f79ba3b31e82c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 08:51:39 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 22 08:51:39 2020 -0700" }, "message": "Merge tag \u0027powerpc-5.7-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux\n\nPull powerpc fixes from Michael Ellerman:\n\n - a revert of a recent change to the PTE bits for 32-bit BookS, which\n broke swap.\n\n - a \"fix\" to disable STRICT_KERNEL_RWX for 64-bit in Kconfig, as it\u0027s\n causing crashes for some people.\n\nThanks to Christophe Leroy and Rui Salvaterra.\n\n* tag \u0027powerpc-5.7-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:\n powerpc/64s: Disable STRICT_KERNEL_RWX\n Revert \"powerpc/32s: reorder Linux PTE bits to better match Hash PTE bits.\"\n" }, { "commit": "7a839dbab1be59f5ed3b3b046de29e166784c9b4", "tree": "70e1058a613930d67ad5c5e10e480944dae9c15b", "parents": [ "1ea34b2988554af8a83a44a0a8e5aaa9e5a60c64" ], "author": { "name": "Klaus Doth", "email": "kdlnx@doth.eu", "time": "Fri May 22 12:56:04 2020 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Fri May 22 13:27:38 2020 +0200" }, "message": "misc: rtsx: Add short delay after exit from ASPM\n\nDMA transfers to and from the SD card stall for 10 seconds and run into\ntimeout on RTS5260 card readers after ASPM was enabled.\n\nAdding a short msleep after disabling ASPM fixes the issue on several\nDell Precision 7530/7540 systems I tested.\n\nThis function is only called when waking up after the chip went into\npower-save after not transferring data for a few seconds. The added\nmsleep does therefore not change anything in data transfer speed or\ninduce any excessive waiting while data transfers are running, or the\nchip is sleeping. Only the transition from sleep to active is affected.\n\nSigned-off-by: Klaus Doth \u003ckdlnx@doth.eu\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nLink: https://lore.kernel.org/r/4434eaa7-2ee3-a560-faee-6cee63ebd6d4@doth.eu\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "5cf65922bb15279402e1e19b5ee8c51d618fa51f", "tree": "1e1ec83eaedcf0e2e973f9cf32d37de8fb8e3264", "parents": [ "dfeb376dd4cb2c5004aeb625e2475f58a5ff2ea7" ], "author": { "name": "Jakub Sitnicki", "email": "jakub@cloudflare.com", "time": "Thu May 21 10:34:35 2020 +0200" }, "committer": { "name": "Alexei Starovoitov", "email": "ast@kernel.org", "time": "Thu May 21 17:52:45 2020 -0700" }, "message": "flow_dissector: Drop BPF flow dissector prog ref on netns cleanup\n\nWhen attaching a flow dissector program to a network namespace with\nbpf(BPF_PROG_ATTACH, ...) we grab a reference to bpf_prog.\n\nIf netns gets destroyed while a flow dissector is still attached, and there\nare no other references to the prog, we leak the reference and the program\nremains loaded.\n\nLeak can be reproduced by running flow dissector tests from selftests/bpf:\n\n # bpftool prog list\n # ./test_flow_dissector.sh\n ...\n selftests: test_flow_dissector [PASS]\n # bpftool prog list\n 4: flow_dissector name _dissect tag e314084d332a5338 gpl\n loaded_at 2020-05-20T18:50:53+0200 uid 0\n xlated 552B jited 355B memlock 4096B map_ids 3,4\n btf_id 4\n #\n\nFix it by detaching the flow dissector program when netns is going away.\n\nFixes: d58e468b1112 (\"flow_dissector: implements flow dissector BPF hook\")\nSigned-off-by: Jakub Sitnicki \u003cjakub@cloudflare.com\u003e\nSigned-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nReviewed-by: Stanislav Fomichev \u003csdf@google.com\u003e\nLink: https://lore.kernel.org/bpf/20200521083435.560256-1-jakub@cloudflare.com\n" }, { "commit": "7d9ff5eed4146bf026c69e766ff630bc0bd555bb", "tree": "d03ac7513ff8b984d86b1d036c769c07741a3da5", "parents": [ "9bf430396900bfb3642eb753c6d38954e9e92716", "31ecebee9c36d5e5e113a357a655d993fa916174" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 22 10:30:50 2020 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 22 10:30:51 2020 +1000" }, "message": "Merge tag \u0027amd-drm-fixes-5.7-2020-05-21\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-fixes\n\namd-drm-fixes-5.7-2020-05-21:\n\namdgpu:\n- DP fix\n- Floating point fix\n- Fix cursor stutter issue\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Alex Deucher \u003calexdeucher@gmail.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20200521222415.4122-1-alexander.deucher@amd.com\n" }, { "commit": "a7654211d0ffeaa8eb0545ea00f8445242cbce05", "tree": "1035541121ddc67b58de30c1f632a07f8926d89a", "parents": [ "41b4bd986f86331efc599b9a3f5fb86ad92e9af9" ], "author": { "name": "Tang Bin", "email": "tangbin@cmss.chinamobile.com", "time": "Wed May 20 17:55:32 2020 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:26:54 2020 -0700" }, "message": "net: sgi: ioc3-eth: Fix return value check in ioc3eth_probe()\n\nIn the function devm_platform_ioremap_resource(), if get resource\nfailed, the return value is ERR_PTR() not NULL. Thus it must be\nreplaced by IS_ERR(), or else it may result in crashes if a critical\nerror path is encountered.\n\nFixes: 0ce5ebd24d25 (\"mfd: ioc3: Add driver for SGI IOC3 chip\")\nSigned-off-by: Zhang Shengju \u003czhangshengju@cmss.chinamobile.com\u003e\nSigned-off-by: Tang Bin \u003ctangbin@cmss.chinamobile.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "41b4bd986f86331efc599b9a3f5fb86ad92e9af9", "tree": "2f289dd4806f0da96746971fd814c120e43e5d21", "parents": [ "57ebc8f08504f176eb0f25b3e0fde517dec61a4f" ], "author": { "name": "Sabrina Dubroca", "email": "sd@queasysnail.net", "time": "Wed May 20 11:15:46 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:25:50 2020 -0700" }, "message": "net: don\u0027t return invalid table id error when we fall back to PF_UNSPEC\n\nIn case we can\u0027t find a -\u003edumpit callback for the requested\n(family,type) pair, we fall back to (PF_UNSPEC,type). In effect, we\u0027re\nin the same situation as if userspace had requested a PF_UNSPEC\ndump. For RTM_GETROUTE, that handler is rtnl_dump_all, which calls all\nthe registered RTM_GETROUTE handlers.\n\nThe requested table id may or may not exist for all of those\nfamilies. commit ae677bbb4441 (\"net: Don\u0027t return invalid table id\nerror when dumping all families\") fixed the problem when userspace\nexplicitly requests a PF_UNSPEC dump, but missed the fallback case.\n\nFor example, when we pass ipv6.disable\u003d1 to a kernel with\nCONFIG_IP_MROUTE\u003dy and CONFIG_IP_MROUTE_MULTIPLE_TABLES\u003dy,\nthe (PF_INET6, RTM_GETROUTE) handler isn\u0027t registered, so we end up in\nrtnl_dump_all, and listing IPv6 routes will unexpectedly print:\n\n # ip -6 r\n Error: ipv4: MR table does not exist.\n Dump terminated\n\ncommit ae677bbb4441 introduced the dump_all_families variable, which\ngets set when userspace requests a PF_UNSPEC dump. However, we can\u0027t\nsimply set the family to PF_UNSPEC in rtnetlink_rcv_msg in the\nfallback case to get dump_all_families \u003d\u003d true, because some messages\ntypes (for example RTM_GETRULE and RTM_GETNEIGH) only register the\nPF_UNSPEC handler and use the family to filter in the kernel what is\ndumped to userspace. We would then export more entries, that userspace\nwould have to filter. iproute does that, but other programs may not.\n\nInstead, this patch removes dump_all_families and updates the\nRTM_GETROUTE handlers to check if the family that is being dumped is\ntheir own. When it\u0027s not, which covers both the intentional PF_UNSPEC\ndumps (as dump_all_families did) and the fallback case, ignore the\nmissing table id error.\n\nFixes: cb167893f41e (\"net: Plumb support for filtering ipv4 and ipv6 multicast route dumps\")\nSigned-off-by: Sabrina Dubroca \u003csd@queasysnail.net\u003e\nReviewed-by: David Ahern \u003cdsahern@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9bf430396900bfb3642eb753c6d38954e9e92716", "tree": "3a760dff4714dbab24dec693b66fadd380ad09c8", "parents": [ "5a3f610877e9d08968ea7237551049581f02b163", "ad99cb5e783bb03d512092db3387ead9504aad3d" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 22 10:24:19 2020 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 22 10:25:11 2020 +1000" }, "message": "Merge branch \u0027etnaviv/fixes\u0027 of https://git.pengutronix.de/git/lst/linux into drm-fixes\n\ntwo fixes:\n- memory leak fix when userspace passes a invalid softpin address\n- off-by-one crashing the kernel in the perfmon domain iteration when\nthe GPU core has both 2D and 3D capabilities\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/ca7bce5c8aff0fcbbdc3bf2b9723df5f687c8924.camel@pengutronix.de\n" }, { "commit": "57ebc8f08504f176eb0f25b3e0fde517dec61a4f", "tree": "0796bf1df2f289283a573456be9195f93af070ea", "parents": [ "a5534617007a1c33095db35c36b6455ee7add00b" ], "author": { "name": "Vadim Fedorenko", "email": "vfedorenko@novek.ru", "time": "Wed May 20 11:50:48 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:24:25 2020 -0700" }, "message": "net: ipip: fix wrong address family in init error path\n\nIn case of error with MPLS support the code is misusing AF_INET\ninstead of AF_MPLS.\n\nFixes: 1b69e7e6c4da (\"ipip: support MPLS over IPv4\")\nSigned-off-by: Vadim Fedorenko \u003cvfedorenko@novek.ru\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a5534617007a1c33095db35c36b6455ee7add00b", "tree": "57d5ed24720738b99f0af664b99cb6648bbc2aa9", "parents": [ "04ba6b7dd8cadeb8ffdc4cba144024c3dad6a6ac", "635d9398178659d8ddba79dd061f9451cec0b4d1" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:20:06 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:20:25 2020 -0700" }, "message": "Merge branch \u0027net-tls-fix-encryption-error-path\u0027\n\nVadim Fedorenko says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet/tls: fix encryption error path\n\nThe problem with data stream corruption was found in KTLS\ntransmit path with small socket send buffers and large\namount of data. bpf_exec_tx_verdict() frees open record\non any type of error including EAGAIN, ENOMEM and ENOSPC\nwhile callers are able to recover this transient errors.\nAlso wrong error code was returned to user space in that\ncase. This patchset fixes the problems.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nAcked-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "635d9398178659d8ddba79dd061f9451cec0b4d1", "tree": "57d5ed24720738b99f0af664b99cb6648bbc2aa9", "parents": [ "a7bff11f6f9afa87c25711db8050c9b5324db0e2" ], "author": { "name": "Vadim Fedorenko", "email": "vfedorenko@novek.ru", "time": "Wed May 20 11:41:44 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:20:06 2020 -0700" }, "message": "net/tls: free record only on encryption error\n\nWe cannot free record on any transient error because it leads to\nlosing previos data. Check socket error to know whether record must\nbe freed or not.\n\nFixes: d10523d0b3d7 (\"net/tls: free the record on encryption error\")\nSigned-off-by: Vadim Fedorenko \u003cvfedorenko@novek.ru\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a7bff11f6f9afa87c25711db8050c9b5324db0e2", "tree": "bb6a3e2bd6a20d7a822094a4998e9c34c897f206", "parents": [ "04ba6b7dd8cadeb8ffdc4cba144024c3dad6a6ac" ], "author": { "name": "Vadim Fedorenko", "email": "vfedorenko@novek.ru", "time": "Wed May 20 11:41:43 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:20:06 2020 -0700" }, "message": "net/tls: fix encryption error checking\n\nbpf_exec_tx_verdict() can return negative value for copied\nvariable. In that case this value will be pushed back to caller\nand the real error code will be lost. Fix it using signed type and\nchecking for positive value.\n\nFixes: d10523d0b3d7 (\"net/tls: free the record on encryption error\")\nFixes: d3b18ad31f93 (\"tls: add bpf support to sk_msg handling\")\nSigned-off-by: Vadim Fedorenko \u003cvfedorenko@novek.ru\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "04ba6b7dd8cadeb8ffdc4cba144024c3dad6a6ac", "tree": "341f692f29bf8bc3c19ab133fd3a09951d259653", "parents": [ "d28ea1fbbf437054ef339afec241019f2c4e2bb6", "1401cf600d548c72f51e20b5841c330d5c11c9e2" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:14:18 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:14:18 2020 -0700" }, "message": "Merge branch \u0027net-ethernet-ti-fix-some-return-value-check\u0027\n\nWei Yongjun says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: ethernet: ti: fix some return value check\n\nThis patchset convert cpsw_ale_create() to return PTR_ERR() only, and\nchanged all the caller to check IS_ERR() instead of NULL.\n\nSince v2:\n1) rebased on net.git, as Jakub\u0027s suggest\n2) split am65-cpsw-nuss.c changes, as Grygorii\u0027s suggest\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": "1401cf600d548c72f51e20b5841c330d5c11c9e2", "tree": "341f692f29bf8bc3c19ab133fd3a09951d259653", "parents": [ "3469660d1b15ccfdf7b33295c306b6298ca730aa" ], "author": { "name": "Wei Yongjun", "email": "weiyongjun1@huawei.com", "time": "Wed May 20 11:41:16 2020 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:14:18 2020 -0700" }, "message": "net: ethernet: ti: am65-cpsw-nuss: fix error handling of am65_cpsw_nuss_probe\n\nConvert to using IS_ERR() instead of NULL test for cpsw_ale_create()\nerror handling. Also fix to return negative error code from this error\nhandling case instead of 0 in.\n\nFixes: 93a76530316a (\"net: ethernet: ti: introduce am65x/j721e gigabit eth subsystem driver\")\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nSigned-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3469660d1b15ccfdf7b33295c306b6298ca730aa", "tree": "965116c1da5b2d6c02070e8e4211b1e436cb61ab", "parents": [ "d28ea1fbbf437054ef339afec241019f2c4e2bb6" ], "author": { "name": "Wei Yongjun", "email": "weiyongjun1@huawei.com", "time": "Wed May 20 11:41:15 2020 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:14:18 2020 -0700" }, "message": "net: ethernet: ti: fix some return value check of cpsw_ale_create()\n\ncpsw_ale_create() can return both NULL and PTR_ERR(), but all of\nthe caller only check NULL for error handling. This patch convert\nit to only return PTR_ERR() in all error cases, and the caller using\nIS_ERR() instead of NULL test.\n\nFixes: 4b41d3436796 (\"net: ethernet: ti: cpsw: allow untagged traffic on host port\")\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nSigned-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d28ea1fbbf437054ef339afec241019f2c4e2bb6", "tree": "d9fad8e096e79975fafe84815dc8fbbd5dac9219", "parents": [ "7c87e32d2e380228ada79d20ac5b7674718ef097" ], "author": { "name": "Manivannan Sadhasivam", "email": "manivannan.sadhasivam@linaro.org", "time": "Tue May 19 23:44:16 2020 +0530" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 17:04:53 2020 -0700" }, "message": "net: qrtr: Fix passing invalid reference to qrtr_local_enqueue()\n\nOnce the traversal of the list is completed with list_for_each_entry(),\nthe iterator (node) will point to an invalid object. So passing this to\nqrtr_local_enqueue() which is outside of the iterator block is erroneous\neventhough the object is not used.\n\nSo fix this by passing NULL to qrtr_local_enqueue().\n\nFixes: bdabad3e363d (\"net: Add Qualcomm IPC router\")\nReported-by: kbuild test robot \u003clkp@intel.com\u003e\nReported-by: Julia Lawall \u003cjulia.lawall@lip6.fr\u003e\nSigned-off-by: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\nReviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7c87e32d2e380228ada79d20ac5b7674718ef097", "tree": "6fb9d504829a14148eecfa46671c2e093deff951", "parents": [ "d69100b8eee27c2d60ee52df76e0b80a8d492d34" ], "author": { "name": "Michal Kubecek", "email": "mkubecek@suse.cz", "time": "Sun May 10 21:04:09 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 21 16:59:19 2020 -0700" }, "message": "ethtool: count header size in reply size estimate\n\nAs ethnl_request_ops::reply_size handlers do not include common header\nsize into calculated/estimated reply size, it needs to be added in\nethnl_default_doit() and ethnl_default_notify() before allocating the\nmessage. On the other hand, strset_reply_size() should not add common\nheader size.\n\nFixes: 728480f12442 (\"ethtool: default handlers for GET requests\")\nReported-by: Oleksij Rempel \u003co.rempel@pengutronix.de\u003e\nSigned-off-by: Michal Kubecek \u003cmkubecek@suse.cz\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "051143e1602d90ea71887d92363edd539d411de5", "tree": "5642155146d071ddff81231c15a277cc48a8e78e", "parents": [ "907fa893258ba6076f5fff32900a461decb9e8c5", "c54d481d71c6849e044690d3960aaebc730224cc" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 21 16:43:35 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 21 16:43:35 2020 -0700" }, "message": "Merge tag \u0027apparmor-pr-2020-05-21\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jj/linux-apparmor\n\nPull apparmor bug fixes from John Johansen:\n\n - Fix use-after-free in aa_audit_rule_init\n\n - Fix refcnt leak in policy_update\n\n - Fix potential label refcnt leak in aa_change_profile\n\n* tag \u0027apparmor-pr-2020-05-21\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jj/linux-apparmor:\n apparmor: Fix use-after-free in aa_audit_rule_init\n apparmor: Fix aa_label refcnt leak in policy_update\n apparmor: fix potential label refcnt leak in aa_change_profile\n" }, { "commit": "907fa893258ba6076f5fff32900a461decb9e8c5", "tree": "27f58a388f4d1841f3c301806f367a5b61666240", "parents": [ "d2f8825ab78e4c18686f3e1a756a30255bb00bf3" ], "author": { "name": "Namjae Jeon", "email": "namjae.jeon@samsung.com", "time": "Fri May 22 08:10:10 2020 +0900" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 21 16:40:11 2020 -0700" }, "message": "exfat: add the dummy mount options to be backward compatible with staging/exfat\n\nAs Ubuntu and Fedora release new version used kernel version equal to or\nhigher than v5.4, They started to support kernel exfat filesystem.\n\nLinus reported a mount error with new version of exfat on Fedora:\n\n exfat: Unknown parameter \u0027namecase\u0027\n\nThis is because there is a difference in mount option between old\nstaging/exfat and new exfat. And utf8, debug, and codepage options as\nwell as namecase have been removed from new exfat.\n\nThis patch add the dummy mount options as deprecated option to be\nbackward compatible with old one.\n\nReported-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nSigned-off-by: Namjae Jeon \u003cnamjae.jeon@samsung.com\u003e\nCc: Matthew Wilcox \u003cwilly@infradead.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Eric Sandeen \u003csandeen@sandeen.net\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c54d481d71c6849e044690d3960aaebc730224cc", "tree": "d43fe9636bb5c992caead0e29a7aba02c2b3dbec", "parents": [ "c6b39f070722ea9963ffe756bfe94e89218c5e63" ], "author": { "name": "Navid Emamdoost", "email": "navid.emamdoost@gmail.com", "time": "Mon Oct 21 10:23:47 2019 -0500" }, "committer": { "name": "John Johansen", "email": "john.johansen@canonical.com", "time": "Thu May 21 15:25:51 2020 -0700" }, "message": "apparmor: Fix use-after-free in aa_audit_rule_init\n\nIn the implementation of aa_audit_rule_init(), when aa_label_parse()\nfails the allocated memory for rule is released using\naa_audit_rule_free(). But after this release, the return statement\ntries to access the label field of the rule which results in\nuse-after-free. Before releasing the rule, copy errNo and return it\nafter release.\n\nFixes: 52e8c38001d8 (\"apparmor: Fix memory leak of rule on error exit path\")\nSigned-off-by: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\nSigned-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n" } ], "next": "c6b39f070722ea9963ffe756bfe94e89218c5e63" }