)]}' { "log": [ { "commit": "6f48655facfd7f7ccfe6d252ac0fe319ab02e4dd", "tree": "d0720bfe2f1552c3fdda2a47438263a67e359b76", "parents": [ "978d13d60c34818a41fc35962602bdfa5c03f214" ], "author": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Aug 06 16:10:03 2017 -0700" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Wed Aug 09 20:55:19 2017 -0700" }, "message": "target: Fix node_acl demo-mode + uncached dynamic shutdown regression\n\nThis patch fixes a generate_node_acls \u003d 1 + cache_dynamic_acls \u003d 0\nregression, that was introduced by\n\n commit 01d4d673558985d9a118e1e05026633c3e2ade9b\n Author: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n Date: Wed Dec 7 12:55:54 2016 -0800\n\nwhich originally had the proper list_del_init() usage, but was\ndropped during list review as it was thought unnecessary by HCH.\n\nHowever, list_del_init() usage is required during the special\ngenerate_node_acls \u003d 1 + cache_dynamic_acls \u003d 0 case when\ntransport_free_session() does a list_del(\u0026se_nacl-\u003eacl_list),\nfollowed by target_complete_nacl() doing the same thing.\n\nThis was manifesting as a general protection fault as reported\nby Justin:\n\nkernel: general protection fault: 0000 [#1] SMP\nkernel: Modules linked in:\nkernel: CPU: 0 PID: 11047 Comm: iscsi_ttx Not tainted 4.13.0-rc2.x86_64.1+ #20\nkernel: Hardware name: Intel Corporation S5500BC/S5500BC, BIOS S5500.86B.01.00.0064.050520141428 05/05/2014\nkernel: task: ffff88026939e800 task.stack: ffffc90007884000\nkernel: RIP: 0010:target_put_nacl+0x49/0xb0\nkernel: RSP: 0018:ffffc90007887d70 EFLAGS: 00010246\nkernel: RAX: dead000000000200 RBX: ffff8802556ca000 RCX: 0000000000000000\nkernel: RDX: dead000000000100 RSI: 0000000000000246 RDI: ffff8802556ce028\nkernel: RBP: ffffc90007887d88 R08: 0000000000000001 R09: 0000000000000000\nkernel: R10: ffffc90007887df8 R11: ffffea0009986900 R12: ffff8802556ce020\nkernel: R13: ffff8802556ce028 R14: ffff8802556ce028 R15: ffffffff88d85540\nkernel: FS: 0000000000000000(0000) GS:ffff88027fc00000(0000) knlGS:0000000000000000\nkernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nkernel: CR2: 00007fffe36f5f94 CR3: 0000000009209000 CR4: 00000000003406f0\nkernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\nkernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\nkernel: Call Trace:\nkernel: transport_free_session+0x67/0x140\nkernel: transport_deregister_session+0x7a/0xc0\nkernel: iscsit_close_session+0x92/0x210\nkernel: iscsit_close_connection+0x5f9/0x840\nkernel: iscsit_take_action_for_connection_exit+0xfe/0x110\nkernel: iscsi_target_tx_thread+0x140/0x1e0\nkernel: ? wait_woken+0x90/0x90\nkernel: kthread+0x124/0x160\nkernel: ? iscsit_thread_get_cpumask+0x90/0x90\nkernel: ? kthread_create_on_node+0x40/0x40\nkernel: ret_from_fork+0x22/0x30\nkernel: Code: 00 48 89 fb 4c 8b a7 48 01 00 00 74 68 4d 8d 6c 24 08 4c\n89 ef e8 e8 28 43 00 48 8b 93 20 04 00 00 48 8b 83 28 04 00 00 4c 89\nef \u003c48\u003e 89 42 08 48 89 10 48 b8 00 01 00 00 00 00 ad de 48 89 83 20\nkernel: RIP: target_put_nacl+0x49/0xb0 RSP: ffffc90007887d70\nkernel: ---[ end trace f12821adbfd46fed ]---\n\nTo address this, go ahead and use proper list_del_list() for all\ncases of se_nacl-\u003eacl_list deletion.\n\nReported-by: Justin Maggard \u003cjmaggard01@gmail.com\u003e\nTested-by: Justin Maggard \u003cjmaggard01@gmail.com\u003e\nCc: Justin Maggard \u003cjmaggard01@gmail.com\u003e\nCc: stable@vger.kernel.org # 4.1+\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "978d13d60c34818a41fc35962602bdfa5c03f214", "tree": "c93d2586f98f345336957007ceda4dcd28605212", "parents": [ "6bcbb3174caa5f1ccc894f8ae077631659d5a629" ], "author": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Fri Aug 04 23:59:31 2017 -0700" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Aug 06 14:41:41 2017 -0700" }, "message": "iscsi-target: Fix iscsi_np reset hung task during parallel delete\n\nThis patch fixes a bug associated with iscsit_reset_np_thread()\nthat can occur during parallel configfs rmdir of a single iscsi_np\nused across multiple iscsi-target instances, that would result in\nhung task(s) similar to below where configfs rmdir process context\nwas blocked indefinately waiting for iscsi_np-\u003enp_restart_comp\nto finish:\n\n[ 6726.112076] INFO: task dcp_proxy_node_:15550 blocked for more than 120 seconds.\n[ 6726.119440] Tainted: G W O 4.1.26-3321 #2\n[ 6726.125045] \"echo 0 \u003e /proc/sys/kernel/hung_task_timeout_secs\" disables this message.\n[ 6726.132927] dcp_proxy_node_ D ffff8803f202bc88 0 15550 1 0x00000000\n[ 6726.140058] ffff8803f202bc88 ffff88085c64d960 ffff88083b3b1ad0 ffff88087fffeb08\n[ 6726.147593] ffff8803f202c000 7fffffffffffffff ffff88083f459c28 ffff88083b3b1ad0\n[ 6726.155132] ffff88035373c100 ffff8803f202bca8 ffffffff8168ced2 ffff8803f202bcb8\n[ 6726.162667] Call Trace:\n[ 6726.165150] [\u003cffffffff8168ced2\u003e] schedule+0x32/0x80\n[ 6726.170156] [\u003cffffffff8168f5b4\u003e] schedule_timeout+0x214/0x290\n[ 6726.176030] [\u003cffffffff810caef2\u003e] ? __send_signal+0x52/0x4a0\n[ 6726.181728] [\u003cffffffff8168d7d6\u003e] wait_for_completion+0x96/0x100\n[ 6726.187774] [\u003cffffffff810e7c80\u003e] ? wake_up_state+0x10/0x10\n[ 6726.193395] [\u003cffffffffa035d6e2\u003e] iscsit_reset_np_thread+0x62/0xe0 [iscsi_target_mod]\n[ 6726.201278] [\u003cffffffffa0355d86\u003e] iscsit_tpg_disable_portal_group+0x96/0x190 [iscsi_target_mod]\n[ 6726.210033] [\u003cffffffffa0363f7f\u003e] lio_target_tpg_store_enable+0x4f/0xc0 [iscsi_target_mod]\n[ 6726.218351] [\u003cffffffff81260c5a\u003e] configfs_write_file+0xaa/0x110\n[ 6726.224392] [\u003cffffffff811ea364\u003e] vfs_write+0xa4/0x1b0\n[ 6726.229576] [\u003cffffffff811eb111\u003e] SyS_write+0x41/0xb0\n[ 6726.234659] [\u003cffffffff8169042e\u003e] system_call_fastpath+0x12/0x71\n\nIt would happen because each iscsit_reset_np_thread() sets state\nto ISCSI_NP_THREAD_RESET, sends SIGINT, and then blocks waiting\nfor completion on iscsi_np-\u003enp_restart_comp.\n\nHowever, if iscsi_np was active processing a login request and\nmore than a single iscsit_reset_np_thread() caller to the same\niscsi_np was blocked on iscsi_np-\u003enp_restart_comp, iscsi_np\nkthread process context in __iscsi_target_login_thread() would\nflush pending signals and only perform a single completion of\nnp-\u003enp_restart_comp before going back to sleep within transport\nspecific iscsit_transport-\u003eiscsi_accept_np code.\n\nTo address this bug, add a iscsi_np-\u003enp_reset_count and update\n__iscsi_target_login_thread() to keep completing np-\u003enp_restart_comp\nuntil -\u003enp_reset_count has reached zero.\n\nReported-by: Gary Guo \u003cghg@datera.io\u003e\nTested-by: Gary Guo \u003cghg@datera.io\u003e\nCc: Mike Christie \u003cmchristi@redhat.com\u003e\nCc: Hannes Reinecke \u003chare@suse.de\u003e\nCc: stable@vger.kernel.org # 3.10+\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "6bcbb3174caa5f1ccc894f8ae077631659d5a629", "tree": "230908352d46945e5d7096cb46cba8fcd83a20d4", "parents": [ "d96adb9b076a12d30500347e2e667689062f44a0" ], "author": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Fri Jun 30 00:08:13 2017 -0700" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Jul 30 16:48:43 2017 -0700" }, "message": "qla2xxx: Fix incorrect tcm_qla2xxx_free_cmd use during TMR ABORT (v2)\n\nThis patch drops two incorrect usages of tcm_qla2xxx_free_cmd()\nduring TMR ABORT within tcm_qla2xxx_handle_data_work() and\ntcm_qla2xxx_aborted_task(), which where attempting to dispatch\ninto workqueue context to do tcm_qla2xxx_complete_free() and\nsubsequently invoke transport_generic_free_cmd().\n\nThis is incorrect because during TMR ABORT target-core will\ndrop the outstanding se_cmd-\u003ecmd_kref references once it has\nquiesced the se_cmd via transport_wait_for_tasks(), and in\nthe case of qla2xxx it should not attempt to do it\u0027s own\ntransport_generic_free_cmd() once the abort has occured.\n\nAs reported by Pascal, this was originally manifesting as a\nBUG_ON(cmd-\u003ecmd_in_wq) in qlt_free_cmd() during TMR ABORT,\nwith a LIO backend that had sufficently high enough WRITE\nlatency to trigger a host side TMR ABORT_TASK.\n\n(v2: Drop the qla_tgt_cmd-\u003ewrite_pending_abort_comp changes,\n as they will be addressed in a seperate series)\n\nReported-by: Pascal de Bruijn \u003cp.debruijn@unilogic.nl\u003e\nTested-by: Pascal de Bruijn \u003cp.debruijn@unilogic.nl\u003e\nCc: Pascal de Bruijn \u003cp.debruijn@unilogic.nl\u003e\nReported-by: Lukasz Engel \u003clukasz.engel@softax.pl\u003e\nCc: Lukasz Engel \u003clukasz.engel@softax.pl\u003e\nAcked-by: Himanshu Madhani \u003chimanshu.madhani@cavium.com\u003e\nCc: Quinn Tran \u003cquinn.tran@cavium.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # 3.10+\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "d96adb9b076a12d30500347e2e667689062f44a0", "tree": "f9c294b8c5824a197899a712869eb425a1c8c688", "parents": [ "310d40a973c560a24c79f84cb5f16dc540a05686" ], "author": { "name": "Varun Prakash", "email": "varun@chelsio.com", "time": "Sat Jul 29 21:01:49 2017 +0530" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Jul 30 15:35:34 2017 -0700" }, "message": "cxgbit: fix sg_nents calculation\n\nThe current logic of calculating sg_nents can fail\nif data_offset % PAGE_SIZE is not zero.\n\nFor example -\nPAGE_SIZE \u003d 4096\ndata_len \u003d 3072\ndata_offset \u003d 3072\n\nAs per current logic\nsg_nents \u003d max(1UL, DIV_ROUND_UP(data_len, PAGE_SIZE));\nsg_nents \u003d max(1UL, DIV_ROUND_UP(3072, 4096));\nsg_nents \u003d 1\n\nBut as data_offset % PAGE_SIZE \u003d 3072 we should skip 3072 bytes\nskip \u003d 3K\nsg_nents \u003d max(1UL, DIV_ROUND_UP(3K(skip) + 3K(data_len), 4K(PAGE_SIZE));\nsg_nents \u003d 2;\n\nThis patch fixes this issue by adding skip to data_len.\n\nSigned-off-by: Varun Prakash \u003cvarun@chelsio.com\u003e\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "310d40a973c560a24c79f84cb5f16dc540a05686", "tree": "c0532c34b8c0a1261aa8cbeed3a6a3b4c1b71ee6", "parents": [ "ea8dc5b4cd2195ee582cae28afa4164c6dea1738" ], "author": { "name": "Varun Prakash", "email": "varun@chelsio.com", "time": "Sun Jul 23 20:03:45 2017 +0530" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Jul 30 15:32:48 2017 -0700" }, "message": "iscsi-target: fix invalid flags in text response\n\nIn case of multiple text responses iscsi-target\nsets both \u0027F\u0027 and \u0027C\u0027 bit for the final text response\npdu, this issue happens because hdr-\u003eflags is not\nzeroed out before ORing with \u0027F\u0027 bit.\n\nThis patch removes the | operator to fix this issue.\n\nSigned-off-by: Varun Prakash \u003cvarun@chelsio.com\u003e\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "ea8dc5b4cd2195ee582cae28afa4164c6dea1738", "tree": "55b2a54d2ccb2213d5cf633306b7ccf9fd8dfda6", "parents": [ "66b59f9b1f41ee61eb4862bbcfc1e7db5298ba9e" ], "author": { "name": "Varun Prakash", "email": "varun@chelsio.com", "time": "Sun Jul 23 20:03:33 2017 +0530" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Jul 30 15:31:34 2017 -0700" }, "message": "iscsi-target: fix memory leak in iscsit_setup_text_cmd()\n\nOn receiving text request iscsi-target allocates buffer for\npayload in iscsit_handle_text_cmd() and assigns buffer pointer\nto cmd-\u003etext_in_ptr, this buffer is currently freed in\niscsit_release_cmd(), if iscsi-target sets \u0027C\u0027 bit in text\nresponse then it will receive another text request from the\ninitiator with ttt !\u003d 0xffffffff in this case iscsi-target\nwill find cmd using itt and call iscsit_setup_text_cmd()\nwhich will set cmd-\u003etext_in_ptr to NULL without freeing\npreviously allocated buffer.\n\nThis patch fixes this issue by calling kfree(cmd-\u003etext_in_ptr)\nin iscsit_setup_text_cmd() before assigning NULL to it.\n\nFor the first text request cmd-\u003etext_in_ptr is NULL as\ncmd is memset to 0 in iscsit_allocate_cmd().\n\nSigned-off-by: Varun Prakash \u003cvarun@chelsio.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # 4.0+\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "66b59f9b1f41ee61eb4862bbcfc1e7db5298ba9e", "tree": "812c7fb23ffbe8c3b7d8bdf5ba152cf4f0b82639", "parents": [ "ededd039d1b96035b23592c049efcae53922cfce" ], "author": { "name": "Varun Prakash", "email": "varun@chelsio.com", "time": "Sun Jul 23 20:03:07 2017 +0530" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Jul 30 15:26:00 2017 -0700" }, "message": "cxgbit: add missing __kfree_skb()\n\nCall __kfree_skb() after processing skb to avoid\nmemory leak.\n\nSigned-off-by: Varun Prakash \u003cvarun@chelsio.com\u003e\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "ededd039d1b96035b23592c049efcae53922cfce", "tree": "36eeb68d6d9fcf2c2ac8ea0e8c35c62fc2c4d172", "parents": [ "c542942cb42186f99b6d715a833c7afad359f48f" ], "author": { "name": "Bryant G. Ly", "email": "bryantly@linux.vnet.ibm.com", "time": "Fri Jul 14 08:11:04 2017 -0500" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Jul 30 15:23:21 2017 -0700" }, "message": "tcmu: free old string on reconfig\n\nOn initial tcmu_configure_device call the info-\u003ename would\nhave already been allocated and set, so on the second call\nmake sure to free it first.\n\nReported-by: Mike Christie \u003cmchristi@redhat.com\u003e\nReviewed-by: Mike Christie \u003cmchristi@redhat.com\u003e\nSigned-off-by: Bryant G. Ly \u003cbryantly@linux.vnet.ibm.com\u003e\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "c542942cb42186f99b6d715a833c7afad359f48f", "tree": "935be87483fc8dcd9029bae5cc61f322fe505121", "parents": [ "520eccdfe187591a51ea9ab4c1a024ae4d0f68d9" ], "author": { "name": "Xiubo Li", "email": "lixiubo@cmss.chinamobile.com", "time": "Wed Jul 12 15:51:17 2017 +0800" }, "committer": { "name": "Nicholas Bellinger", "email": "nab@linux-iscsi.org", "time": "Sun Jul 30 15:12:32 2017 -0700" }, "message": "tcmu: Fix possible to/from address overflow when doing the memcpy\n\nFor most case the sg-\u003elength equals to PAGE_SIZE, so this bug won\u0027t\nbe triggered. Otherwise this will crash the kernel, for example when\nall segments\u0027 sg-\u003elength equal to 1K.\n\nSigned-off-by: Xiubo Li \u003clixiubo@cmss.chinamobile.com\u003e\nReviewed-by: Mike Christie \u003cmchristi@redhat.com\u003e\nSigned-off-by: Nicholas Bellinger \u003cnab@linux-iscsi.org\u003e\n" }, { "commit": "520eccdfe187591a51ea9ab4c1a024ae4d0f68d9", "tree": "70d61c9156db9ee1f1c6311c9e8421570b8bfc2b", "parents": [ "7683e9e529258d01ce99216ad3be21f59eff83ec" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 23 16:15:17 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 23 16:15:17 2017 -0700" }, "message": "Linux 4.13-rc2\n" }, { "commit": "7683e9e529258d01ce99216ad3be21f59eff83ec", "tree": "bec6de4820826f60cdb866e3e1ae724f12144edd", "parents": [ "f47e07bc5f1a5c48ed60a8ee55352cb4b2bf4d51" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 23 16:06:21 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 23 16:06:21 2017 -0700" }, "message": "Properly alphabetize MAINTAINERS file\n\nThis adds a perl script to actually parse the MAINTAINERS file, clean up\nsome whitespace in it, warn about errors in it, and then properly sort\nthe end result.\n\nMy perl-fu is atrocious, so the script has basically been created by\nrandomly putting various characters in a pile, mixing them around, and\nthen looking it the end result does anything interesting when used as a\nperl script.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f47e07bc5f1a5c48ed60a8ee55352cb4b2bf4d51", "tree": "accec60441cab6f9f19ca20e1855a1b9c5b5020a", "parents": [ "a56e88ec05df50110f2bf578b6e17128f37111ed" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 23 15:08:05 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 23 15:08:05 2017 -0700" }, "message": "Fix up MAINTAINERS file problems\n\nPrepping for scripting the MAINTAINERS file cleanup (and possible split)\nshowed a couple of cases where the headers for a couple of entries were\nbogus.\n\nThere\u0027s a few different kinds of bogosities:\n\n - the X-GENE SOC EDAC case was confused and split over two lines\n\n - there were four entries for \"GREYBUS PROTOCOLS DRIVERS\" that were all\n different things.\n\n - the NOKIA N900 CAMERA SUPPORT\" was duplicated\n\nall of which were more obvious when you started doing associative arrays\nin perl to track these things by the header (so that we can alphabetize\nthis thing properly, and so that we might split it up by the data too).\n\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a56e88ec05df50110f2bf578b6e17128f37111ed", "tree": "fe3ee0bcd1cef2d28ffd62a0c012ad972c581049", "parents": [ "4b162c530d9c101381500e586fedb1340595a6ff", "96edd61dcf44362d3ef0bed1a5361e0ac7886a63" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 23 11:22:45 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 23 11:22:45 2017 -0700" }, "message": "Merge tag \u0027for-linus-4.13b-rc2-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip\n\nPull xen fixes from Juergen Gross:\n \"Some fixes and cleanups for running under Xen\"\n\n* tag \u0027for-linus-4.13b-rc2-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:\n xen/balloon: don\u0027t online new memory initially\n xen/x86: fix cpu hotplug\n xen/grant-table: log the lack of grants\n xen/x86: Don\u0027t BUG on CPU0 offlining\n" }, { "commit": "96edd61dcf44362d3ef0bed1a5361e0ac7886a63", "tree": "fe3ee0bcd1cef2d28ffd62a0c012ad972c581049", "parents": [ "c185ddec54657c145a0c2055e4b87918da24974f" ], "author": { "name": "Juergen Gross", "email": "jgross@suse.com", "time": "Mon Jul 10 10:10:45 2017 +0200" }, "committer": { "name": "Juergen Gross", "email": "jgross@suse.com", "time": "Sun Jul 23 08:13:18 2017 +0200" }, "message": "xen/balloon: don\u0027t online new memory initially\n\nWhen setting up the Xenstore watch for the memory target size the new\nwatch will fire at once. Don\u0027t try to reach the configured target size\nby onlining new memory in this case, as the current memory size will\nbe smaller in almost all cases due to e.g. BIOS reserved pages.\n\nOnlining new memory will lead to more problems e.g. undesired conflicts\nwith NVMe devices meant to be operated as block devices.\n\nInstead remember the difference between target size and current size\nwhen the watch fires for the first time and apply it to any further\nsize changes, too.\n\nIn order to avoid races between balloon.c and xen-balloon.c init calls\ndo the xen-balloon.c initialization from balloon.c.\n\nSigned-off-by: Juergen Gross \u003cjgross@suse.com\u003e\nReviewed-by: Boris Ostrovsky \u003cboris.ostrovsky@oracle.com\u003e\nSigned-off-by: Juergen Gross \u003cjgross@suse.com\u003e\n" }, { "commit": "c185ddec54657c145a0c2055e4b87918da24974f", "tree": "534b075a9ed9871aed3983cfaa86c7111310b69d", "parents": [ "29d11cfd8698038b87458ba4d1329b9da81150a5" ], "author": { "name": "Juergen Gross", "email": "jgross@suse.com", "time": "Wed Jul 05 16:05:20 2017 +0200" }, "committer": { "name": "Juergen Gross", "email": "jgross@suse.com", "time": "Sun Jul 23 08:13:11 2017 +0200" }, "message": "xen/x86: fix cpu hotplug\n\nCommit dc6416f1d711eb4c1726e845d653235dcaae12e1 (\"xen/x86: Call\ncpu_startup_entry(CPUHP_AP_ONLINE_IDLE) from xen_play_dead()\")\nintroduced an error leading to a stack overflow of the idle task when\na cpu was brought offline/online many times: by calling\ncpu_startup_entry() instead of returning at the end of xen_play_dead()\ndo_idle() would be entered again and again.\n\nDon\u0027t use cpu_startup_entry(), but cpuhp_online_idle() instead allowing\nto return from xen_play_dead().\n\nCc: \u003cstable@vger.kernel.org\u003e # 4.12\nSigned-off-by: Juergen Gross \u003cjgross@suse.com\u003e\nReviewed-by: Boris Ostrovsky \u003cboris.ostrovsky@oracle.com\u003e\nSigned-off-by: Juergen Gross \u003cjgross@suse.com\u003e\n" }, { "commit": "29d11cfd8698038b87458ba4d1329b9da81150a5", "tree": "916c500fd8617537daafa63975da37b9d5f713b9", "parents": [ "0e4d394fe50308f69eb061eb89de78279df1e9dc" ], "author": { "name": "Wengang Wang", "email": "wen.gang.wang@oracle.com", "time": "Tue Jul 18 09:40:35 2017 +0200" }, "committer": { "name": "Juergen Gross", "email": "jgross@suse.com", "time": "Sun Jul 23 08:09:43 2017 +0200" }, "message": "xen/grant-table: log the lack of grants\n\nlog a message when we enter this situation:\n1) we already allocated the max number of available grants from hypervisor\nand\n2) we still need more (but the request fails because of 1)).\n\nSometimes the lack of grants causes IO hangs in xen_blkfront devices.\nAdding this log would help debuging.\n\nSigned-off-by: Wengang Wang \u003cwen.gang.wang@oracle.com\u003e\nReviewed-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nReviewed-by: Junxiao Bi \u003cjunxiao.bi@oracle.com\u003e\nReviewed-by: Juergen Gross \u003cjgross@suse.com\u003e\nSigned-off-by: Juergen Gross \u003cjgross@suse.com\u003e\n" }, { "commit": "0e4d394fe50308f69eb061eb89de78279df1e9dc", "tree": "1ded5a4769dd411d8da8ac61e0388bcb956d4e6b", "parents": [ "4b162c530d9c101381500e586fedb1340595a6ff" ], "author": { "name": "Vitaly Kuznetsov", "email": "vkuznets@redhat.com", "time": "Mon Jun 26 18:39:30 2017 +0200" }, "committer": { "name": "Juergen Gross", "email": "jgross@suse.com", "time": "Sun Jul 23 08:09:24 2017 +0200" }, "message": "xen/x86: Don\u0027t BUG on CPU0 offlining\n\nCONFIG_BOOTPARAM_HOTPLUG_CPU0 allows to offline CPU0 but Xen HVM guests\nBUG() in xen_teardown_timer(). Remove the BUG_ON(), this is probably a\nleftover from ancient times when CPU0 hotplug was impossible, it works\njust fine for HVM.\n\nSigned-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nAcked-by: Juergen Gross \u003cjgross@suse.com\u003e\nSigned-off-by: Juergen Gross \u003cjgross@suse.com\u003e\n" }, { "commit": "4b162c530d9c101381500e586fedb1340595a6ff", "tree": "51057a35a53fe53a03e246a38b2ac7daef9cf899", "parents": [ "ae75d1aefe04c30de9b344abc69939cf7b537ac6", "1009ccdc64ee2c8451f76b548589f6b989d13412" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 09:25:00 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 09:25:00 2017 -0700" }, "message": "Merge tag \u0027hwmon-for-linus-v4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging\n\nPull hwmon fix from Guenter Roeck:\n \"Avoid buffer overruns in applesmc driver\"\n\n* tag \u0027hwmon-for-linus-v4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:\n hwmon: (applesmc) Avoid buffer overruns\n" }, { "commit": "ae75d1aefe04c30de9b344abc69939cf7b537ac6", "tree": "c4309b781624ae92cc17e043ed1c1b93215231db", "parents": [ "dedaff2f6d0dd1e88022da6aaa3dbc97f7b5cf8b", "c6325179238f1d4683edbec53d8322575d76d7e2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 09:00:24 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 09:00:24 2017 -0700" }, "message": "Merge tag \u0027tty-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty\n\nPull tty/serial fixes from Greg KH:\n \"Here are some small tty and serial driver fixes for 4.13-rc2. Nothing\n huge at all, a revert of a patch that turned out to break things, a\n fix up for a new tty ioctl we added in 4.13-rc1 to get the uapi\n definition correct, and a few minor serial driver fixes for reported\n issues.\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027tty-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:\n tty: Fix TIOCGPTPEER ioctl definition\n tty: hide unused pty_get_peer function\n tty: serial: lpuart: Fix the logic for detecting the 32-bit type UART\n serial: imx: Prevent TX buffer PIO write when a DMA has been started\n Revert \"serial: imx-serial - move DMA buffer configuration to DT\"\n serial: sh-sci: Uninitialized variables in sysfs files\n serial: st-asc: Potential error pointer dereference\n" }, { "commit": "dedaff2f6d0dd1e88022da6aaa3dbc97f7b5cf8b", "tree": "774b9d5fe9398e7deca930ff979346f6a150c9fd", "parents": [ "55fd939e8ee4538c3b66a28975dcc3beb96f80ba", "c89876dda01841a6a485cb29b9d1843db34958a3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 08:57:24 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 08:57:24 2017 -0700" }, "message": "Merge tag \u0027char-misc-4.13-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc\n\nPull char/misc driver fixes from Greg KH:\n \"Here are some small char and misc driver fixes for 4.13-rc2. All fix\n reported problems with 4.13-rc1 or older kernels (like the binder\n fixes). Full details in the shortlog.\n\n All have been in linux-next with no reported issues\"\n\n* tag \u0027char-misc-4.13-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:\n w1: omap-hdq: fix error return code in omap_hdq_probe()\n regmap: regmap-w1: Fix build troubles\n w1: Fix slave count on 1-Wire bus (resend)\n mux: mux-core: unregister mux_class in mux_exit()\n mux: remove the Kconfig question for the subsystem\n nvmem: rockchip-efuse: amend compatible rk322x-efuse to rk3228-efuse\n drivers/fsi: fix fsi_slave_mode prototype\n fsi: core: register with postcore_initcall\n thunderbolt: Correct access permissions for active NVM contents\n vmbus: re-enable channel tasklet\n spmi: pmic-arb: Always allocate ppid_to_apid table\n MAINTAINERS: Add entry for SPMI subsystem\n spmi: Include OF based modalias in device uevent\n binder: Use wake up hint for synchronous transactions.\n binder: use group leader instead of open thread\n Revert \"android: binder: Sanity check at binder ioctl\"\n" }, { "commit": "55fd939e8ee4538c3b66a28975dcc3beb96f80ba", "tree": "d378a35decdb2ba4e5d0f90579342a7a0bdaf87f", "parents": [ "bcb53e5769227813b7878df1ec9d329b0bd68f74", "d6f5f071f1e13cadecf8aef1faa7e5d6fbc9f33b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 08:55:16 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 08:55:16 2017 -0700" }, "message": "Merge tag \u0027usb-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nPull USB fixes from Greg KH:\n \"Here are some small USB fixes for 4.13-rc2.\n\n The usual batch, gadget fixes for reported issues, as well as xhci\n fixes, and a small random collection of other fixes for reported\n issues.\n\n All have been in linux-next with no reported issues\"\n\n* tag \u0027usb-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (25 commits)\n xhci: fix memleak in xhci_run()\n usb: xhci: fix spinlock recursion for USB2 test mode\n xhci: fix 20000ms port resume timeout\n usb: xhci: Issue stop EP command only when the EP state is running\n xhci: Bad Ethernet performance plugged in ASM1042A host\n xhci: Fix NULL pointer dereference when cleaning up streams for removed host\n usb: renesas_usbhs: gadget: disable all eps when the driver stops\n usb: renesas_usbhs: fix usbhsc_resume() for !USBHSF_RUNTIME_PWCTRL\n usb: gadget: udc: renesas_usb3: protect usb3_ep-\u003estarted in usb3_start_pipen()\n usb: gadget: udc: renesas_usb3: fix zlp transfer by the dmac\n usb: gadget: udc: renesas_usb3: fix free size in renesas_usb3_dma_free_prd()\n usb: gadget: f_uac2: endianness fixes.\n usb: gadget: f_uac1: endianness fixes.\n include: usb: audio: specify exact endiannes of descriptors\n usb: gadget: udc: start_udc() can be static\n usb: dwc2: gadget: On USB RESET reset device address to zero\n usb: storage: return on error to avoid a null pointer dereference\n usb: typec: include linux/device.h in ucsi.h\n USB: cdc-acm: add device-id for quirky printer\n usb: dwc3: gadget: only unmap requests from DMA if mapped\n ...\n" }, { "commit": "bcb53e5769227813b7878df1ec9d329b0bd68f74", "tree": "96b374131cf6bc4bae12dfaea97bea707200c341", "parents": [ "82abbea734d659b4218ad06734b4927b43261985", "5a1d4c5dd4eb2f1f8a9b30e61762f3b3b564df70" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 08:53:24 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 22 08:53:24 2017 -0700" }, "message": "Merge tag \u0027staging-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging\n\nPull staging driver fixes from Greg KH:\n \"Here are some small staging driver fixes for reported issues for\n 4.13-rc2.\n\n Also in here is a new driver, the virtualbox DRM driver. It\u0027s\n stand-alone and got acks from the DRM developers to go in through this\n tree. It\u0027s a new thing, but it should be fine for this point in the rc\n cycle due to it being independent.\n\n All of this has been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027staging-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:\n staging: rtl8188eu: add TL-WN722N v2 support\n staging: speakup: safely register and unregister ldisc\n staging: speakup: add functions to register and unregister ldisc\n staging: speakup: safely close tty\n staging: sm750fb: avoid conflicting vesafb\n staging: lustre: ko2iblnd: check copy_from_iter/copy_to_iter return code\n staging: vboxvideo: Add vboxvideo to drivers/staging\n staging: sm750fb: fixed a assignment typo\n staging: rtl8188eu: memory leak in rtw_free_cmd_obj()\n staging: vchiq_arm: fix error codes in probe\n staging: comedi: ni_mio_common: fix AO timer off-by-one regression\n" }, { "commit": "82abbea734d659b4218ad06734b4927b43261985", "tree": "7be60ecbf6e1911e53c686d19b5ec56c627d7cfd", "parents": [ "505d5c11192960a3f0639d1d9e05dffeddd4e874" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@infradead.org", "time": "Fri Jul 21 13:32:27 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 18:59:11 2017 -0700" }, "message": "MAINTAINERS: fix alphabetical ordering\n\nFix major alphabetic errors. No attempt to fix items that all begin\nwith the same word (like ARM, BROADCOM, DRM, EDAC, FREESCALE, INTEL,\nOMAP, PCI, SAMSUNG, TI, USB, etc.).\n\n(diffstat +/- is different by one line because TI KEYSTONE MULTICORE\nhad 2 blank lines after it.)\n\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nAcked-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "505d5c11192960a3f0639d1d9e05dffeddd4e874", "tree": "4b9f151bda0f7c146107851bb9053fc1ec4c6461", "parents": [ "99313414dda451a8c00b139afc77a803e647f309", "1ebf980127924c639e2b85c08468311ba1c95b70" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 16:26:01 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 16:26:01 2017 -0700" }, "message": "Merge tag \u0027nfs-for-4.13-2\u0027 of git://git.linux-nfs.org/projects/anna/linux-nfs\n\nPull NFS client bugfixes from Anna Schumaker:\n \"Stable bugfix:\n - Fix error reporting regression\n\n Bugfixes:\n - Fix setting filelayout ds address race\n - Fix subtle access bug when using ACLs\n - Fix setting mnt3_counts array size\n - Fix a couple of pNFS commit races\"\n\n* tag \u0027nfs-for-4.13-2\u0027 of git://git.linux-nfs.org/projects/anna/linux-nfs:\n NFS/filelayout: Fix racy setting of fl-\u003edsaddr in filelayout_check_deviceid()\n NFS: Be more careful about mapping file permissions\n NFS: Store the raw NFS access mask in the inode\u0027s access cache\n NFSv3: Convert nfs3_proc_access() to use nfs_access_set_mask()\n NFS: Refactor NFS access to kernel access mask calculation\n net/sunrpc/xprt_sock: fix regression in connection error reporting.\n nfs: count correct array for mnt3_counts array size\n Revert commit 722f0b891198 (\"pNFS: Don\u0027t send COMMITs to the DSes if...\")\n pNFS/flexfiles: Handle expired layout segments in ff_layout_initiate_commit()\n NFS: Fix another COMMIT race in pNFS\n NFS: Fix a COMMIT race in pNFS\n mount: copy the port field into the cloned nfs_server structure.\n NFS: Don\u0027t run wake_up_bit() when nobody is waiting...\n nfs: add export operations\n" }, { "commit": "99313414dda451a8c00b139afc77a803e647f309", "tree": "8f23042ffbbd231154c6eceb469fcb9c941c4b7f", "parents": [ "0151ef0085f946eed75da851297577f88d74c50c", "0e082555cec9510d276965fe391f709acb32c0f4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 16:24:22 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 16:24:22 2017 -0700" }, "message": "Merge branch \u0027overlayfs-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs\n\nPull overlayfs fixes from Miklos Szeredi:\n \"This fixes a crash with SELinux and several other old and new bugs\"\n\n* \u0027overlayfs-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:\n ovl: check for bad and whiteout index on lookup\n ovl: do not cleanup directory and whiteout index entries\n ovl: fix xattr get and set with selinux\n ovl: remove unneeded check for IS_ERR()\n ovl: fix origin verification of index dir\n ovl: mark parent impure on ovl_link()\n ovl: fix random return value on mount\n" }, { "commit": "0151ef0085f946eed75da851297577f88d74c50c", "tree": "9ab2f69450555848a5c4d0d473f06b248e1a07c2", "parents": [ "bb236dbeea8181c6733e6d34a53bfef9c8ef4e95", "42de82a8b544fa55670feef7d6f85085fba48fc0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 16:20:05 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 16:20:05 2017 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.dk/linux-block\n\nPull block fixes from Jens Axboe:\n \"A small set of fixes for -rc2 - two fixes for BFQ, documentation and\n code, and a removal of an unused variable in nbd. Outside of that, a\n small collection of fixes from the usual crew on the nvme side\"\n\n* \u0027for-linus\u0027 of git://git.kernel.dk/linux-block:\n nvmet: don\u0027t report 0-bytes in serial number\n nvmet: preserve controller serial number between reboots\n nvmet: Move serial number from controller to subsystem\n nvmet: prefix version configfs file with attr\n nvme-pci: Fix an error handling path in \u0027nvme_probe()\u0027\n nvme-pci: Remove nvme_setup_prps BUG_ON\n nvme-pci: add another device ID with stripe quirk\n nvmet-fc: fix byte swapping in nvmet_fc_ls_create_association\n nvme: fix byte swapping in the streams code\n nbd: kill unused ret in recv_work\n bfq: dispatch request to prevent queue stalling after the request completion\n bfq: fix typos in comments about B-WF2Q+ algorithm\n" }, { "commit": "bb236dbeea8181c6733e6d34a53bfef9c8ef4e95", "tree": "2d71f98dcc5676beb4e6029462d60be0ff6ebbc1", "parents": [ "24a1635a41bccb5cc426eaef8b88c7e0961ef6bb", "a62ab66b13a0f9bcb17b7b761f6670941ed5cd62" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 14:22:05 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 14:22:05 2017 -0700" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma\n\nPull more rdma fixes from Doug Ledford:\n \"As per my previous pull request, there were two drivers that each had\n a rather large number of legitimate fixes still to be sent.\n\n As it turned out, I also missed a reasonably large set of fixes from\n one person across the stack that are all important fixes. All in all,\n the bnxt_re, i40iw, and Dan Carpenter are 3/4 to 2/3rds of this pull\n request.\n\n There were some other random fixes that I didn\u0027t send in the last pull\n request that I added to this one. This catches the rdma stack up to\n the fixes from up to about the beginning of this week. Any more fixes\n I\u0027ll wait and batch up later in the -rc cycle. This will give us a\n good base to start with for basing a for-next branch on -rc2.\n\n Summary:\n\n - i40iw fixes\n\n - bnxt_re fixes\n\n - Dan Carpenter bugfixes across stack\n\n - ten more random fixes, no more than two from any one person\"\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (37 commits)\n RDMA/core: Initialize port_num in qp_attr\n RDMA/uverbs: Fix the check for port number\n IB/cma: Fix reference count leak when no ipv4 addresses are set\n RDMA/iser: don\u0027t send an rkey if all data is written as immadiate-data\n rxe: fix broken receive queue draining\n RDMA/qedr: Prevent memory overrun in verbs\u0027 user responses\n iw_cxgb4: don\u0027t use WR keys/addrs for 0 byte reads\n IB/mlx4: Fix CM REQ retries in paravirt mode\n IB/rdmavt: Setting of QP timeout can overflow jiffies computation\n IB/core: Fix sparse warnings\n RDMA/bnxt_re: Fix the value reported for local ack delay\n RDMA/bnxt_re: Report MISSED_EVENTS in req_notify_cq\n RDMA/bnxt_re: Fix return value of poll routine\n RDMA/bnxt_re: Enable atomics only if host bios supports\n RDMA/bnxt_re: Specify RDMA component when allocating stats context\n RDMA/bnxt_re: Fixed the max_rd_atomic support for initiator and destination QP\n RDMA/bnxt_re: Report supported value to IB stack in query_device\n RDMA/bnxt_re: Do not free the ctx_tbl entry if delete GID fails\n RDMA/bnxt_re: Fix WQE Size posted to HW to prevent it from throwing error\n RDMA/bnxt_re: Free doorbell page index (DPI) during dealloc ucontext\n ...\n" }, { "commit": "24a1635a41bccb5cc426eaef8b88c7e0961ef6bb", "tree": "eefc2f6d4a2bacd056d119d9331447e227bf5652", "parents": [ "f79ec886f994918de057dd224fa2dfdc162bb3c3", "5896ec77d70d33dd38a455b0aa5f3154aeecea09" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 14:16:42 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 14:16:42 2017 -0700" }, "message": "Merge tag \u0027drm-fixes-for-v4.13-rc2\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull drm fixes from Dave Airlie:\n \"A bunch of fixes for rc2: two imx regressions, vc4 fix, dma-buf fix,\n some displayport mst fixes, and an amdkfd fix.\n\n Nothing too crazy, I assume we just haven\u0027t see much rc1 testing yet\"\n\n* tag \u0027drm-fixes-for-v4.13-rc2\u0027 of git://people.freedesktop.org/~airlied/linux:\n drm/mst: Avoid processing partially received up/down message transactions\n drm/mst: Avoid dereferencing a NULL mstb in drm_dp_mst_handle_up_req()\n drm/mst: Fix error handling during MST sideband message reception\n drm/imx: parallel-display: Accept drm_of_find_panel_or_bridge failure\n drm/imx: fix typo in ipu_plane_formats[]\n drm/vc4: Fix VBLANK handling in crtc-\u003eenable() path\n dma-buf/fence: Avoid use of uninitialised timestamp\n drm/amdgpu: Remove unused field kgd2kfd_shared_resources.num_mec\n drm/radeon: Remove initialization of shared_resources.num_mec\n drm/amdkfd: Remove unused references to shared_resources.num_mec\n drm/amdgpu: Fix KFD oversubscription by tracking queues correctly\n" }, { "commit": "f79ec886f994918de057dd224fa2dfdc162bb3c3", "tree": "8d66a5ebf3d2f9475a908b7c5b37675965ee1c8d", "parents": [ "b0a752818bdba3d0fe9de593618ccab9f1101508", "f86f418059b94aa01f9342611a272ca60c583e89" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 13:59:51 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 13:59:51 2017 -0700" }, "message": "Merge tag \u0027trace-v4.13-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing fixes from Steven Rostedt:\n \"Three minor updates\n\n - Use the new GFP_RETRY_MAYFAIL to be more aggressive in allocating\n memory for the ring buffer without causing OOMs\n\n - Fix a memory leak in adding and removing instances\n\n - Add __rcu annotation to be able to debug RCU usage of function\n tracing a bit better\"\n\n* tag \u0027trace-v4.13-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n trace: fix the errors caused by incompatible type of RCU variables\n tracing: Fix kmemleak in instance_rmdir\n tracing/ring_buffer: Try harder to allocate\n" }, { "commit": "b0a752818bdba3d0fe9de593618ccab9f1101508", "tree": "e0e664be7583415ed4e7db41710c9691c0f27128", "parents": [ "10fc95547fcc672ef1fc18ae1b70e1e9f52f246a", "f1ff89ec4447c4e39d275a1ca3de43eed2a92745" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 13:58:10 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 13:58:10 2017 -0700" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm\n\nPull KVM fixes from Radim Krčmář:\n \"A bunch of small fixes for x86\"\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm:\n kvm: x86: hyperv: avoid livelock in oneshot SynIC timers\n KVM: VMX: Fix invalid guest state detection after task-switch emulation\n x86: add MULTIUSER dependency for KVM\n KVM: nVMX: Disallow VM-entry in MOV-SS shadow\n KVM: nVMX: track NMI blocking state separately for each VMCS\n KVM: x86: masking out upper bits\n" }, { "commit": "10fc95547fcc672ef1fc18ae1b70e1e9f52f246a", "tree": "f8ea61061326061debe3f1472c1312159e39a674", "parents": [ "4ec9f7a18b9fcef6e8f7c13279b48e3bb5d4d704", "029d9252b116fa52a95150819e62af1f6e420fe5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 13:54:37 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 13:54:37 2017 -0700" }, "message": "Merge tag \u0027powerpc-4.13-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux\n\nPull powerpc fixes from Michael Ellerman:\n \"A handful of fixes, mostly for new code:\n\n - some reworking of the new STRICT_KERNEL_RWX support to make sure we\n also remove executable permission from __init memory before it\u0027s\n freed.\n\n - a fix to some recent optimisations to the hypercall entry where we\n were clobbering r12, this was breaking nested guests (PR KVM).\n\n - a fix for the recent patch to opal_configure_cores(). This could\n break booting on bare metal Power8 boxes if the kernel was built\n without CONFIG_JUMP_LABEL_FEATURE_CHECK_DEBUG.\n\n - .. and finally a workaround for spurious PMU interrupts on Power9\n DD2.\n\n Thanks to: Nicholas Piggin, Anton Blanchard, Balbir Singh\"\n\n* tag \u0027powerpc-4.13-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:\n powerpc/mm: Mark __init memory no-execute when STRICT_KERNEL_RWX\u003dy\n powerpc/mm/hash: Refactor hash__mark_rodata_ro()\n powerpc/mm/radix: Refactor radix__mark_rodata_ro()\n powerpc/64s: Fix hypercall entry clobbering r12 input\n powerpc/perf: Avoid spurious PMU interrupts after idle\n powerpc/powernv: Fix boot on Power8 bare metal due to opal_configure_cores()\n" }, { "commit": "4ec9f7a18b9fcef6e8f7c13279b48e3bb5d4d704", "tree": "b1b1b49ff1e9b483ab03e66549523eca9985659d", "parents": [ "e234b4a807202990a0aec4e74aa552fe76b5b6d2", "db15e7f27369b81b6605a546d54eb844f87370a5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:20:58 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:20:58 2017 -0700" }, "message": "Merge branch \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Ingo Molnar:\n \"Half of the fixes are for various build time warnings triggered by\n randconfig builds. Most (but not all...) were harmless.\n\n There\u0027s also:\n\n - ACPI boundary condition fixes\n\n - UV platform fixes\n\n - defconfig updates\n\n - an AMD K6 CPU init fix\n\n - a %pOF printk format related preparatory change\n\n - .. and a warning fix related to the tlb/PCID changes\"\n\n* \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/devicetree: Convert to using %pOF instead of -\u003efull_name\n x86/platform/uv/BAU: Disable BAU on single hub configurations\n x86/platform/intel-mid: Fix a format string overflow warning\n x86/platform: Add PCI dependency for PUNIT_ATOM_DEBUG\n x86/build: Silence the build with \"make -s\"\n x86/io: Add \"memory\" clobber to insb/insw/insl/outsb/outsw/outsl\n x86/fpu/math-emu: Avoid bogus -Wint-in-bool-context warning\n x86/fpu/math-emu: Fix possible uninitialized variable use\n perf/x86: Shut up false-positive -Wmaybe-uninitialized warning\n x86/defconfig: Remove stale, old Kconfig options\n x86/ioapic: Pass the correct data to unmask_ioapic_irq()\n x86/acpi: Prevent out of bound access caused by broken ACPI tables\n x86/mm, KVM: Fix warning when !CONFIG_PREEMPT_COUNT\n x86/platform/uv/BAU: Fix congested_response_us not taking effect\n x86/cpu: Use indirect call to measure performance in init_amd_k6()\n" }, { "commit": "e234b4a807202990a0aec4e74aa552fe76b5b6d2", "tree": "72e23ab12916c8104af8dcf701718a95d86e1af6", "parents": [ "5a77f0254b35e0bea414305e1b71a6bdaee814dc", "32f2fea6e77e64cd4045ec2d5deb879aada3b476" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:18:09 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:18:09 2017 -0700" }, "message": "Merge branch \u0027timers-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull timer fix from Ingo Molnar:\n \"A timer_irq_init() clocksource API robustness fix\"\n\n* \u0027timers-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n clocksource/drivers/timer-of: Handle of_irq_get_byname() result correctly\n" }, { "commit": "5a77f0254b35e0bea414305e1b71a6bdaee814dc", "tree": "b4c91e3e0b57dbd66a9889e762b307d4d093cfb4", "parents": [ "bbcdea658f42070d25e7764f1b81785a51cb1642", "193be41e33168a3a06eb9d356d9e39c69de161d2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:16:12 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:16:12 2017 -0700" }, "message": "Merge branch \u0027sched-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull scheduler fixes from Ingo Molnar:\n \"A cputime fix and code comments/organization fix to the deadline\n scheduler\"\n\n* \u0027sched-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n sched/deadline: Fix confusing comments about selection of top pi-waiter\n sched/cputime: Don\u0027t use smp_processor_id() in preemptible context\n" }, { "commit": "bbcdea658f42070d25e7764f1b81785a51cb1642", "tree": "ea5c21c180560918e94aa83a939c9d2507812ff5", "parents": [ "8b810a3a35eec9cd5e9245266dd0ae10de063094", "df6c3db8d30fb1699ccbc403196b86324f4257af" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:12:48 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:12:48 2017 -0700" }, "message": "Merge branch \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf fixes from Ingo Molnar:\n \"Two hw-enablement patches, two race fixes, three fixes for regressions\n of semantics, plus a number of tooling fixes\"\n\n* \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n perf/x86/intel: Add proper condition to run sched_task callbacks\n perf/core: Fix locking for children siblings group read\n perf/core: Fix scheduling regression of pinned groups\n perf/x86/intel: Fix debug_store reset field for freq events\n perf/x86/intel: Add Goldmont Plus CPU PMU support\n perf/x86/intel: Enable C-state residency events for Apollo Lake\n perf symbols: Accept zero as the kernel base address\n Revert \"perf/core: Drop kernel samples even though :u is specified\"\n perf annotate: Fix broken arrow at row 0 connecting jmp instruction to its target\n perf evsel: State in the default event name if attr.exclude_kernel is set\n perf evsel: Fix attr.exclude_kernel setting for default cycles:p\n" }, { "commit": "8b810a3a35eec9cd5e9245266dd0ae10de063094", "tree": "0aec06b7cca4eeae6c6778f5c23d4f3b1dbc38f1", "parents": [ "34eddefee41a56b378d866897bf83ed909fb4ab3", "69f0d429c413fe96db2c187475cebcc6e3a8c7f5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:11:23 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:11:23 2017 -0700" }, "message": "Merge branch \u0027locking-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull locking fixlet from Ingo Molnar:\n \"Remove an unnecessary priority adjustment in the rtmutex code\"\n\n* \u0027locking-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n locking/rtmutex: Remove unnecessary priority adjustment\n" }, { "commit": "1ebf980127924c639e2b85c08468311ba1c95b70", "tree": "514331ff7257270cc81f6963cc483b91cd0f0530", "parents": [ "ecbb903c56745d59c301db26dd7d8b74b520eb84" ], "author": { "name": "Trond Myklebust", "email": "trond.myklebust@primarydata.com", "time": "Thu Jul 20 17:00:02 2017 -0400" }, "committer": { "name": "Anna Schumaker", "email": "Anna.Schumaker@Netapp.com", "time": "Fri Jul 21 14:08:45 2017 -0400" }, "message": "NFS/filelayout: Fix racy setting of fl-\u003edsaddr in filelayout_check_deviceid()\n\nWe must set fl-\u003edsaddr once, and once only, even if there are multiple\nprocesses calling filelayout_check_deviceid() for the same layout\nsegment.\n\nReported-by: Olga Kornievskaia \u003ckolga@netapp.com\u003e\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\nSigned-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n" }, { "commit": "34eddefee41a56b378d866897bf83ed909fb4ab3", "tree": "80bb3ce474fb0c0f031122313544b0a40e63f6f5", "parents": [ "0a6109fd1b605cbfa58ae9ae9d571d0f4140b727", "acc80c39929b9f2ff8b45fcfe103385a3e45c1a7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:07:41 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 11:07:41 2017 -0700" }, "message": "Merge branch \u0027irq-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull irq fixes from Ingo Molnar:\n \"A resume_irq() fix, plus a number of static declaration fixes\"\n\n* \u0027irq-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n irqchip/digicolor: Drop unnecessary static\n irqchip/mips-cpu: Drop unnecessary static\n irqchip/gic/realview: Drop unnecessary static\n irqchip/mips-gic: Remove population of irq domain names\n genirq/PM: Properly pretend disabled state when force resuming interrupts\n" }, { "commit": "0a6109fd1b605cbfa58ae9ae9d571d0f4140b727", "tree": "ecb3be1ce7bab9fe8821a9c40c8ba22fbee1f32d", "parents": [ "921edf312a6a20be16cf2b60e0dec3dce35e5cb9", "325cdacd03c12629aa5f9ee2ace49b1f3dc184a8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 10:41:19 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 21 10:41:19 2017 -0700" }, "message": "Merge branch \u0027core-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull core fixes from Ingo Molnar:\n \"A fix to WARN_ON_ONCE() done by modules, plus a MAINTAINERS update\"\n\n* \u0027core-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n debug: Fix WARN_ON_ONCE() for modules\n MAINTAINERS: Update the PTRACE entry\n" }, { "commit": "ecbb903c56745d59c301db26dd7d8b74b520eb84", "tree": "882b94947ed7bae985aa9535066452a74d01d404", "parents": [ "bd8b2441742b49c76bec707757bd9c028ea9838e" ], "author": { "name": "Trond Myklebust", "email": "trond.myklebust@primarydata.com", "time": "Tue Jul 11 17:54:35 2017 -0400" }, "committer": { "name": "Anna Schumaker", "email": "Anna.Schumaker@Netapp.com", "time": "Fri Jul 21 11:51:19 2017 -0400" }, "message": "NFS: Be more careful about mapping file permissions\n\nWhen mapping a directory, we want the MAY_WRITE permissions to reflect\nwhether or not we have permission to modify, add and delete the directory\nentries. MAY_EXEC must map to lookup permissions.\n\nOn the other hand, for files, we want MAY_WRITE to reflect a permission\nto modify and extend the file.\n\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\nSigned-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n" }, { "commit": "bd8b2441742b49c76bec707757bd9c028ea9838e", "tree": "950978dc49c18fe92a7e17b63f3a07286c463e1a", "parents": [ "eda3e20847788c453aa7ab478aeaceb56ed29cb6" ], "author": { "name": "Trond Myklebust", "email": "trond.myklebust@primarydata.com", "time": "Tue Jul 11 17:54:34 2017 -0400" }, "committer": { "name": "Anna Schumaker", "email": "Anna.Schumaker@Netapp.com", "time": "Fri Jul 21 11:51:19 2017 -0400" }, "message": "NFS: Store the raw NFS access mask in the inode\u0027s access cache\n\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\nSigned-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n" }, { "commit": "eda3e20847788c453aa7ab478aeaceb56ed29cb6", "tree": "5c3de3389a93be3284932a333853f3de129f84ec", "parents": [ "15d4b73ac2232d6f2beb61d8b2400ea66e4da606" ], "author": { "name": "Trond Myklebust", "email": "trond.myklebust@primarydata.com", "time": "Tue Jul 11 17:54:33 2017 -0400" }, "committer": { "name": "Anna Schumaker", "email": "Anna.Schumaker@Netapp.com", "time": "Fri Jul 21 11:51:19 2017 -0400" }, "message": "NFSv3: Convert nfs3_proc_access() to use nfs_access_set_mask()\n\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\nSigned-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n" }, { "commit": "15d4b73ac2232d6f2beb61d8b2400ea66e4da606", "tree": "18e3f28f6b4e33aed2750283eed0ae5f2116a685", "parents": [ "3ffbc1d65583394be12801655781dd2b079ce169" ], "author": { "name": "Trond Myklebust", "email": "trond.myklebust@primarydata.com", "time": "Tue Jul 11 17:54:32 2017 -0400" }, "committer": { "name": "Anna Schumaker", "email": "Anna.Schumaker@Netapp.com", "time": "Fri Jul 21 11:51:19 2017 -0400" }, "message": "NFS: Refactor NFS access to kernel access mask calculation\n\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\nSigned-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n" }, { "commit": "3ffbc1d65583394be12801655781dd2b079ce169", "tree": "a906fde05bff9b28dc499c89cb9c799867c86107", "parents": [ "ecc7b435d2beb025d7506af74cf749af2cef5734" ], "author": { "name": "NeilBrown", "email": "neilb@suse.com", "time": "Wed Jul 19 14:05:01 2017 +1000" }, "committer": { "name": "Anna Schumaker", "email": "Anna.Schumaker@Netapp.com", "time": "Fri Jul 21 08:49:58 2017 -0400" }, "message": "net/sunrpc/xprt_sock: fix regression in connection error reporting.\n\nCommit 3d4762639dd3 (\"tcp: remove poll() flakes when receiving\nRST\") in v4.12 changed the order in which -\u003esk_state_change()\nand -\u003esk_error_report() are called when a socket is shut\ndown - sk_state_change() is now called first.\n\nThis causes xs_tcp_state_change() -\u003e xs_sock_mark_closed() -\u003e\nxprt_disconnect_done() to wake all pending tasked with -EAGAIN.\nWhen the -\u003esk_error_report() callback arrives, it is too late to\npass the error on, and it is lost.\n\nAs easy way to demonstrate the problem caused is to try to start\nrpc.nfsd while rcpbind isn\u0027t running.\nnfsd will attempt a tcp connection to rpcbind. A ECONNREFUSED\nerror is returned, but sunrpc code loses the error and keeps\nretrying. If it saw the ECONNREFUSED, it would abort.\n\nTo fix this, handle the sk-\u003esk_err in the TCP_CLOSE branch of\nxs_tcp_state_change().\n\nFixes: 3d4762639dd3 (\"tcp: remove poll() flakes when receiving RST\")\nCc: stable@vger.kernel.org (v4.12)\nSigned-off-by: NeilBrown \u003cneilb@suse.com\u003e\nSigned-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n" }, { "commit": "ecc7b435d2beb025d7506af74cf749af2cef5734", "tree": "e8df3a94d1df437bb83f50a332860324fb276351", "parents": [ "213297369cf4900eba906dd32ce845074e30f487" ], "author": { "name": "Eryu Guan", "email": "eguan@redhat.com", "time": "Tue Jul 18 13:32:32 2017 +0800" }, "committer": { "name": "Anna Schumaker", "email": "Anna.Schumaker@Netapp.com", "time": "Fri Jul 21 08:49:57 2017 -0400" }, "message": "nfs: count correct array for mnt3_counts array size\n\nArray size of mnt3_counts should be the size of array\nmnt3_procedures, not mnt_procedures, though they\u0027re same in size\nright now. Found this by code inspection.\n\nFixes: 1c5876ddbdb4 (\"sunrpc: move p_count out of struct rpc_procinfo\")\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Eryu Guan \u003ceguan@redhat.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n" }, { "commit": "db15e7f27369b81b6605a546d54eb844f87370a5", "tree": "b6747ea8c36ab90c3eb96792a84c44922805993a", "parents": [ "2fe9a5c6ade4dfb53ff1c137cca3828d9d1d0948" ], "author": { "name": "Rob Herring", "email": "robh@kernel.org", "time": "Tue Jul 18 16:42:47 2017 -0500" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Fri Jul 21 10:14:15 2017 +0200" }, "message": "x86/devicetree: Convert to using %pOF instead of -\u003efull_name\n\nNow that we have a custom printf format specifier, convert users of\nfull_name to use %pOF instead. This is preparation to remove storing\nof the full path string for each device node.\n\nSigned-off-by: Rob Herring \u003crobh@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: devicetree@vger.kernel.org\nLink: http://lkml.kernel.org/r/20170718214339.7774-7-robh@kernel.org\n[ Clarify the error message while at it, as \u0027node\u0027 is ambiguous. ]\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "df6c3db8d30fb1699ccbc403196b86324f4257af", "tree": "50f70bbeee63d2f4c0ebe3ccea94311a8be8aac0", "parents": [ "2aeb1883547626d82c597cce2c99f0b9c62e2425" ], "author": { "name": "Jiri Olsa", "email": "jolsa@redhat.com", "time": "Wed Jul 19 09:52:47 2017 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Fri Jul 21 09:58:39 2017 +0200" }, "message": "perf/x86/intel: Add proper condition to run sched_task callbacks\n\nWe have 2 functions using the same sched_task callback:\n\n - PEBS drain for free running counters\n - LBR save/store\n\nBoth of them are called from intel_pmu_sched_task() and\neither of them can be unwillingly triggered when the\nother one is configured to run.\n\nLet\u0027s say there\u0027s PEBS drain configured in sched_task\ncallback for the event, but in the callback itself\n(intel_pmu_sched_task()) we will also run the code for\nLBR save/restore, which we did not ask for, but the\ncode in intel_pmu_sched_task() does not check for that.\n\nThis can lead to extra cycles in some perf monitoring,\nlike when we monitor PEBS event without LBR data.\n\n # perf record --no-timestamp -c 10000 -e cycles:p ./perf bench sched pipe -l 1000000\n\n (We need PEBS, non freq/non timestamp event to enable\n the sched_task callback)\n\nThe perf stat of cycles and msr:write_msr for above\ncommand before the change:\n ...\n Performance counter stats for \u0027./perf record --no-timestamp -c 10000 -e cycles:p \\\n ./perf bench sched pipe -l 1000000\u0027 (5 runs):\n\n 18,519,557,441 cycles:k\n 91,195,527 msr:write_msr\n\n 29.334476406 seconds time elapsed\n\nAnd after the change:\n ...\n Performance counter stats for \u0027./perf record --no-timestamp -c 10000 -e cycles:p \\\n ./perf bench sched pipe -l 1000000\u0027 (5 runs):\n\n 18,704,973,540 cycles:k\n 27,184,720 msr:write_msr\n\n 16.977875900 seconds time elapsed\n\nThere\u0027s no affect on cycles:k because the sched_task happens\nwith events switched off, however the msr:write_msr tracepoint\ncounter together with almost 50% of time speedup show the\nimprovement.\n\nMonitoring LBR event and having extra PEBS drain processing\nin sched_task callback showed just a little speedup, because\nthe drain function does not do much extra work in case there\nis no PEBS data.\n\nAdding conditions to recognize the configured work that needs\nto be done in the x86_pmu\u0027s sched_task callback.\n\nSuggested-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nSigned-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\nAcked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: Kan Liang \u003ckan.liang@intel.com\u003e\nCc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nLink: http://lkml.kernel.org/r/20170719075247.GA27506@krava\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "2fe9a5c6ade4dfb53ff1c137cca3828d9d1d0948", "tree": "38f1f1c2fd93e1bc6d04a9211748536b58a390cd", "parents": [ "0bc73048d7baecf94117d1a948853a627e6ba5c8" ], "author": { "name": "Andrew Banman", "email": "abanman@hpe.com", "time": "Thu Jul 20 17:05:51 2017 -0500" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Fri Jul 21 09:56:25 2017 +0200" }, "message": "x86/platform/uv/BAU: Disable BAU on single hub configurations\n\nThe BAU confers no benefit to a UV system running with only one hub/socket.\nPermanently disable the BAU driver if there are less than two hubs online\nto avoid BAU overhead. We have observed failed boots on single-socket UV4\nsystems caused by BAU that are avoided with this patch.\n\nAlso, while at it, consolidate initialization error blocks and fix a\nmemory leak.\n\nSigned-off-by: Andrew Banman \u003cabanman@hpe.com\u003e\nAcked-by: Russ Anderson \u003crja@hpe.com\u003e\nAcked-by: Mike Travis \u003cmike.travis@hpe.com\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: tony.ernst@hpe.com\nLink: http://lkml.kernel.org/r/1500588351-78016-1-git-send-email-abanman@hpe.com\n[ Minor cleanups. ]\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "2aeb1883547626d82c597cce2c99f0b9c62e2425", "tree": "5d3656ca9ae600e7b85f0d040f6b2ce80ccfa363", "parents": [ "3bda69c1c3993a2bddbae01397d12bfef6054011" ], "author": { "name": "Jiri Olsa", "email": "jolsa@kernel.org", "time": "Thu Jul 20 16:14:55 2017 +0200" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Fri Jul 21 09:54:23 2017 +0200" }, "message": "perf/core: Fix locking for children siblings group read\n\nWe\u0027re missing ctx lock when iterating children siblings\nwithin the perf_read path for group reading. Following\nrace and crash can happen:\n\nUser space doing read syscall on event group leader:\n\nT1:\n perf_read\n lock event-\u003ectx-\u003emutex\n perf_read_group\n lock leader-\u003echild_mutex\n __perf_read_group_add(child)\n list_for_each_entry(sub, \u0026leader-\u003esibling_list, group_entry)\n\n----\u003e sub might be invalid at this point, because it could\n get removed via perf_event_exit_task_context in T2\n\nChild exiting and cleaning up its events:\n\nT2:\n perf_event_exit_task_context\n lock ctx-\u003emutex\n list_for_each_entry_safe(child_event, next, \u0026child_ctx-\u003eevent_list,...\n perf_event_exit_event(child)\n lock ctx-\u003elock\n perf_group_detach(child)\n unlock ctx-\u003elock\n\n----\u003e child is removed from sibling_list without any sync\n with T1 path above\n\n ...\n free_event(child)\n\nBefore the child is removed from the leader\u0027s child_list,\n(and thus is omitted from perf_read_group processing), we\nneed to ensure that perf_read_group touches child\u0027s\nsiblings under its ctx-\u003elock.\n\nPeter further notes:\n\n| One additional note; this bug got exposed by commit:\n|\n| ba5213ae6b88 (\"perf/core: Correct event creation with PERF_FORMAT_GROUP\")\n|\n| which made it possible to actually trigger this code-path.\n\nTested-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\nAcked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nCc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nCc: Jiri Olsa \u003cjolsa@redhat.com\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nFixes: ba5213ae6b88 (\"perf/core: Correct event creation with PERF_FORMAT_GROUP\")\nLink: http://lkml.kernel.org/r/20170720141455.2106-1-jolsa@kernel.org\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "5896ec77d70d33dd38a455b0aa5f3154aeecea09", "tree": "6fc39002d3b4d30de3dadc426fa393bfa5e82e27", "parents": [ "660f6b5c639228cf53b4eecda86c8dd9055bed53", "799ee2970485dc206c3bf347d6e6827c04d5e4f9" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 21 14:04:44 2017 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 21 14:04:44 2017 +1000" }, "message": "Merge tag \u0027imx-drm-fixes-2017-07-18\u0027 of git://git.pengutronix.de/git/pza/linux into drm-fixes\n\nimx-drm: fix parallel display regression and typo in plane format list\n\n- Fix a regression where the parallel-display driver would not probe\n anymore if no panel is specified in the device tree, since the\n introduction of drm_of_find_panel_or_bridge.\n- Fix a typo in the plane format list: replace a duplicate BGRA8888 format\n with BGRX8888, as originally intended.\n\n* tag \u0027imx-drm-fixes-2017-07-18\u0027 of git://git.pengutronix.de/git/pza/linux:\n drm/imx: parallel-display: Accept drm_of_find_panel_or_bridge failure\n drm/imx: fix typo in ipu_plane_formats[]\n" }, { "commit": "660f6b5c639228cf53b4eecda86c8dd9055bed53", "tree": "668cf1238b694bfc101f413bd7058383ca92f1ea", "parents": [ "22a548d04853297b3855dcac2881d0cb52e3b99c", "636c4c3e762b62aa93632c645ca65879285b16e3" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 21 14:04:10 2017 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 21 14:04:10 2017 +1000" }, "message": "Merge tag \u0027drm-misc-fixes-2017-07-20\u0027 of git://anongit.freedesktop.org/git/drm-misc into drm-fixes\n\nCore Changes:\n- fence: Introduce new fence flag to signify timestamp is populated (Chris)\n- mst: Avoid processing incomplete data + fix NULL dereference (Imre)\n\nDriver Changes:\n- vc4: Avoid WARN from grabbing a ref from vblank that\u0027s not on (Boris)\n\nCc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: Boris Brezillon \u003cboris.brezillon@free-electrons.com\u003e\nCc: Imre Deak \u003cimre.deak@intel.com\u003e\n\n* tag \u0027drm-misc-fixes-2017-07-20\u0027 of git://anongit.freedesktop.org/git/drm-misc:\n drm/mst: Avoid processing partially received up/down message transactions\n drm/mst: Avoid dereferencing a NULL mstb in drm_dp_mst_handle_up_req()\n drm/mst: Fix error handling during MST sideband message reception\n drm/vc4: Fix VBLANK handling in crtc-\u003eenable() path\n dma-buf/fence: Avoid use of uninitialised timestamp\n" }, { "commit": "22a548d04853297b3855dcac2881d0cb52e3b99c", "tree": "3e7d64542b720f1618b4cf2d125acfa3176b487e", "parents": [ "5771a8c08880cdca3bfb4a3fc6d309d6bba20877", "f835edf9aec13bc8abb3dd4836151773cc359f42" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 21 14:00:40 2017 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 21 14:00:40 2017 +1000" }, "message": "Merge tag \u0027drm-amdkfd-fixes-2017-07-18\u0027 of git://people.freedesktop.org/~gabbayo/linux into drm-fixes\n\nThis is a pull request for 4.13-rc2/3. The first patch addresses a bug that\nappears when doing over-subscription of queues, due to a miscalculation of\nthe number of available compute queues. The other three patches remove\ndead code that was created from the bugfix.\n\n* tag \u0027drm-amdkfd-fixes-2017-07-18\u0027 of git://people.freedesktop.org/~gabbayo/linux:\n drm/amdgpu: Remove unused field kgd2kfd_shared_resources.num_mec\n drm/radeon: Remove initialization of shared_resources.num_mec\n drm/amdkfd: Remove unused references to shared_resources.num_mec\n drm/amdgpu: Fix KFD oversubscription by tracking queues correctly\n" }, { "commit": "921edf312a6a20be16cf2b60e0dec3dce35e5cb9", "tree": "de7a67d798a76de632c56effdf29109fc2af760a", "parents": [ "96080f697786e0a30006fcbcc5b53f350fcb3e9f" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Fri Jul 14 11:25:13 2017 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri Jul 21 04:37:22 2017 +0100" }, "message": "ide: avoid warning for timings calculation\n\ngcc-7 warns about the result of a constant multiplication used as\na boolean:\n\ndrivers/ide/ide-timings.c: In function \u0027ide_timing_quantize\u0027:\ndrivers/ide/ide-timings.c:112:24: error: \u0027*\u0027 in boolean context, suggest \u0027\u0026\u0026\u0027 instead [-Werror\u003dint-in-bool-context]\n q-\u003esetup \u003d EZ(t-\u003esetup * 1000, T);\n\nThis slightly rearranges the macro to simplify the code and avoid\nthe warning at the same time.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "96080f697786e0a30006fcbcc5b53f350fcb3e9f", "tree": "77b130e0e7714a55ce2a579f9524173e151e3b2d", "parents": [ "63a86362130f4c17eaa57f3ef5171ec43111a54e", "cbf5ecb305601d063dc94a57680dfbc3f96c188d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 16:33:39 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 16:33:39 2017 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net\n\nPull networking fixes from David Miller:\n\n 1) BPF verifier signed/unsigned value tracking fix, from Daniel\n Borkmann, Edward Cree, and Josef Bacik.\n\n 2) Fix memory allocation length when setting up calls to\n -\u003endo_set_mac_address, from Cong Wang.\n\n 3) Add a new cxgb4 device ID, from Ganesh Goudar.\n\n 4) Fix FIB refcount handling, we have to set it\u0027s initial value before\n the configure callback (which can bump it). From David Ahern.\n\n 5) Fix double-free in qcom/emac driver, from Timur Tabi.\n\n 6) A bunch of gcc-7 string format overflow warning fixes from Arnd\n Bergmann.\n\n 7) Fix link level headroom tests in ip_do_fragment(), from Vasily\n Averin.\n\n 8) Fix chunk walking in SCTP when iterating over error and parameter\n headers. From Alexander Potapenko.\n\n 9) TCP BBR congestion control fixes from Neal Cardwell.\n\n10) Fix SKB fragment handling in bcmgenet driver, from Doug Berger.\n\n11) BPF_CGROUP_RUN_PROG_SOCK_OPS needs to check for null __sk, from Cong\n Wang.\n\n12) xmit_recursion in ppp driver needs to be per-device not per-cpu,\n from Gao Feng.\n\n13) Cannot release skb-\u003edst in UDP if IP options processing needs it.\n From Paolo Abeni.\n\n14) Some netdev ioctl ifr_name[] NULL termination fixes. From Alexander\n Levin and myself.\n\n15) Revert some rtnetlink notification changes that are causing\n regressions, from David Ahern.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (83 commits)\n net: bonding: Fix transmit load balancing in balance-alb mode\n rds: Make sure updates to cp_send_gen can be observed\n net: ethernet: ti: cpsw: Push the request_irq function to the end of probe\n ipv4: initialize fib_trie prior to register_netdev_notifier call.\n rtnetlink: allocate more memory for dev_set_mac_address()\n net: dsa: b53: Add missing ARL entries for BCM53125\n bpf: more tests for mixed signed and unsigned bounds checks\n bpf: add test for mixed signed and unsigned bounds checks\n bpf: fix up test cases with mixed signed/unsigned bounds\n bpf: allow to specify log level and reduce it for test_verifier\n bpf: fix mixed signed/unsigned derived min/max value bounds\n ipv6: avoid overflow of offset in ip6_find_1stfragopt\n net: tehuti: don\u0027t process data if it has not been copied from userspace\n Revert \"rtnetlink: Do not generate notifications for CHANGEADDR event\"\n net: dsa: mv88e6xxx: Enable CMODE config support for 6390X\n dt-binding: ptp: Add SoC compatibility strings for dte ptp clock\n NET: dwmac: Make dwmac reset unconditional\n net: Zero terminate ifr_name in dev_ifname().\n wireless: wext: terminate ifr name coming from userspace\n netfilter: fix netfilter_net_init() return\n ...\n" }, { "commit": "cbf5ecb305601d063dc94a57680dfbc3f96c188d", "tree": "9d1aff82ddb8f20d9220ce6796c5341d5494dcb8", "parents": [ "e623a48ee433985f6ca0fb238f0002cc2eccdf53" ], "author": { "name": "Kosuke Tatsukawa", "email": "tatsu@ab.jp.nec.com", "time": "Thu Jul 20 05:20:40 2017 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:34:40 2017 -0700" }, "message": "net: bonding: Fix transmit load balancing in balance-alb mode\n\nbalance-alb mode used to have transmit dynamic load balancing feature\nenabled by default. However, transmit dynamic load balancing no longer\nworks in balance-alb after commit 8b426dc54cf4 (\"bonding: remove\nhardcoded value\").\n\nBoth balance-tlb and balance-alb use the function bond_do_alb_xmit() to\nsend packets. This function uses the parameter tlb_dynamic_lb.\ntlb_dynamic_lb used to have the default value of 1 for balance-alb, but\nnow the value is set to 0 except in balance-tlb.\n\nRe-enable transmit dyanmic load balancing by initializing tlb_dynamic_lb\nfor balance-alb similar to balance-tlb.\n\nFixes: 8b426dc54cf4 (\"bonding: remove hardcoded value\")\nSigned-off-by: Kosuke Tatsukawa \u003ctatsu@ab.jp.nec.com\u003e\nAcked-by: Andy Gospodarek \u003candy@greyhouse.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e623a48ee433985f6ca0fb238f0002cc2eccdf53", "tree": "e3dc5acbe886b14e0dde4296b6b78748ac7175cd", "parents": [ "070f9c658a59f9a736b1c040001d37b0952e778e" ], "author": { "name": "Håkon Bugge", "email": "Haakon.Bugge@oracle.com", "time": "Thu Jul 20 12:28:55 2017 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:33:01 2017 -0700" }, "message": "rds: Make sure updates to cp_send_gen can be observed\n\ncp-\u003ecp_send_gen is treated as a normal variable, although it may be\nused by different threads.\n\nThis is fixed by using {READ,WRITE}_ONCE when it is incremented and\nREAD_ONCE when it is read outside the {acquire,release}_in_xmit\nprotection.\n\nNormative reference from the Linux-Kernel Memory Model:\n\n Loads from and stores to shared (but non-atomic) variables should\n be protected with the READ_ONCE(), WRITE_ONCE(), and\n ACCESS_ONCE().\n\nClause 5.1.2.4/25 in the C standard is also relevant.\n\nSigned-off-by: Håkon Bugge \u003chaakon.bugge@oracle.com\u003e\nReviewed-by: Knut Omang \u003cknut.omang@oracle.com\u003e\nAcked-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "070f9c658a59f9a736b1c040001d37b0952e778e", "tree": "1a568073b6ce4688fe696550221d2f8493f01bc0", "parents": [ "8799a221f5944a7d74516ecf46d58c28ec1d1f75" ], "author": { "name": "Keerthy", "email": "j-keerthy@ti.com", "time": "Thu Jul 20 16:59:52 2017 +0530" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:28:54 2017 -0700" }, "message": "net: ethernet: ti: cpsw: Push the request_irq function to the end of probe\n\nPush the request_irq function to the end of probe so as\nto ensure all the required fields are populated in the event\nof an ISR getting executed right after requesting the irq.\n\nCurrently while loading the crash kernel a crash was seen as\nsoon as devm_request_threaded_irq was called. This was due to\nn-\u003epoll being NULL which is called as part of net_rx_action\nfunction.\n\nSuggested-by: Sekhar Nori \u003cnsekhar@ti.com\u003e\nSigned-off-by: Keerthy \u003cj-keerthy@ti.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8799a221f5944a7d74516ecf46d58c28ec1d1f75", "tree": "fd9245101926c80873a577dc18c1de1a118fd448", "parents": [ "153711f9421be5dbc973dc57a4109dc9d54c89b1" ], "author": { "name": "Mahesh Bandewar", "email": "maheshb@google.com", "time": "Wed Jul 19 15:41:33 2017 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:24:45 2017 -0700" }, "message": "ipv4: initialize fib_trie prior to register_netdev_notifier call.\n\nNet stack initialization currently initializes fib-trie after the\nfirst call to netdevice_notifier() call. In fact fib_trie initialization\nneeds to happen before first rtnl_register(). It does not cause any problem\nsince there are no devices UP at this moment, but trying to bring \u0027lo\u0027\nUP at initialization would make this assumption wrong and exposes the issue.\n\nFixes following crash\n\n Call Trace:\n ? alternate_node_alloc+0x76/0xa0\n fib_table_insert+0x1b7/0x4b0\n fib_magic.isra.17+0xea/0x120\n fib_add_ifaddr+0x7b/0x190\n fib_netdev_event+0xc0/0x130\n register_netdevice_notifier+0x1c1/0x1d0\n ip_fib_init+0x72/0x85\n ip_rt_init+0x187/0x1e9\n ip_init+0xe/0x1a\n inet_init+0x171/0x26c\n ? ipv4_offload_init+0x66/0x66\n do_one_initcall+0x43/0x160\n kernel_init_freeable+0x191/0x219\n ? rest_init+0x80/0x80\n kernel_init+0xe/0x150\n ret_from_fork+0x22/0x30\n Code: f6 46 23 04 74 86 4c 89 f7 e8 ae 45 01 00 49 89 c7 4d 85 ff 0f 85 7b ff ff ff 31 db eb 08 4c 89 ff e8 16 47 01 00 48 8b 44 24 38 \u003c45\u003e 8b 6e 14 4d 63 76 74 48 89 04 24 0f 1f 44 00 00 48 83 c4 08\n RIP: kmem_cache_alloc+0xcf/0x1c0 RSP: ffff9b1500017c28\n CR2: 0000000000000014\n\nFixes: 7b1a74fdbb9e (\"[NETNS]: Refactor fib initialization so it can handle multiple namespaces.\")\nFixes: 7f9b80529b8a (\"[IPV4]: fib hash|trie initialization\")\n\nSigned-off-by: Mahesh Bandewar \u003cmaheshb@google.com\u003e\nAcked-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "153711f9421be5dbc973dc57a4109dc9d54c89b1", "tree": "06308fcd9cd16cf966d238f9ee60e6bac1a62e3f", "parents": [ "be35e8c516c1915a3035d266a2015b41f73ba3f9" ], "author": { "name": "WANG Cong", "email": "xiyou.wangcong@gmail.com", "time": "Thu Jul 20 11:27:57 2017 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:23:22 2017 -0700" }, "message": "rtnetlink: allocate more memory for dev_set_mac_address()\n\nvirtnet_set_mac_address() interprets mac address as struct\nsockaddr, but upper layer only allocates dev-\u003eaddr_len\nwhich is ETH_ALEN + sizeof(sa_family_t) in this case.\n\nWe lack a unified definition for mac address, so just fix\nthe upper layer, this also allows drivers to interpret it\nto struct sockaddr freely.\n\nReported-by: David Ahern \u003cdsahern@gmail.com\u003e\nSigned-off-by: Cong Wang \u003cxiyou.wangcong@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "be35e8c516c1915a3035d266a2015b41f73ba3f9", "tree": "a5e783439bca3dc8d8b930b60c75f7d9db432fe6", "parents": [ "5067f4cf23cb073cf664372804104c4a5891dc59" ], "author": { "name": "Florian Fainelli", "email": "f.fainelli@gmail.com", "time": "Thu Jul 20 12:25:22 2017 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:21:54 2017 -0700" }, "message": "net: dsa: b53: Add missing ARL entries for BCM53125\n\nThe BCM53125 entry was missing an arl_entries member which would\nbasically prevent the ARL search from terminating properly. This switch\nhas 4 ARL entries, so add that.\n\nFixes: 1da6df85c6fb (\"net: dsa: b53: Implement ARL add/del/dump operations\")\nSigned-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nReviewed-by: Vivien Didelot \u003cvivien.didelot@savoirfairelinux.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5067f4cf23cb073cf664372804104c4a5891dc59", "tree": "16ade798c25aadabe9df7a5d39ced646dec0170d", "parents": [ "6399f1fae4ec29fab5ec76070435555e256ca3a6", "8641250251bfcd93479c71783c6792ae3325d7e4" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:20:28 2017 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:20:28 2017 -0700" }, "message": "Merge branch \u0027BPF-map-value-adjust-fix\u0027\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\nBPF map value adjust fix\n\nFirst patch in the series is the actual fix and the remaining\npatches are just updates to 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": "8641250251bfcd93479c71783c6792ae3325d7e4", "tree": "16ade798c25aadabe9df7a5d39ced646dec0170d", "parents": [ "b712296a41ce0a114895fdff68fc22aada165b07" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Jul 21 00:00:25 2017 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:20:27 2017 -0700" }, "message": "bpf: more tests for mixed signed and unsigned bounds checks\n\nAdd a couple of more test cases to BPF selftests that are related\nto mixed signed and unsigned checks.\n\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b712296a41ce0a114895fdff68fc22aada165b07", "tree": "388161638f63429c33566ff41b6ef3b5d9f3179a", "parents": [ "a1502132866fd2d2705eef4041dd6d7d849f48a2" ], "author": { "name": "Edward Cree", "email": "ecree@solarflare.com", "time": "Fri Jul 21 00:00:24 2017 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:20:27 2017 -0700" }, "message": "bpf: add test for mixed signed and unsigned bounds checks\n\nThese failed due to a bug in verifier bounds handling.\n\nSigned-off-by: Edward Cree \u003cecree@solarflare.com\u003e\nAcked-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a1502132866fd2d2705eef4041dd6d7d849f48a2", "tree": "8b374fdc3a4f54d303077524b486c35bef76b130", "parents": [ "d655490417ee22da3267fe6592a0ec2023c3c0db" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Jul 21 00:00:23 2017 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:20:27 2017 -0700" }, "message": "bpf: fix up test cases with mixed signed/unsigned bounds\n\nFix the few existing test cases that used mixed signed/unsigned\nbounds and switch them only to one flavor. Reason why we need this\nis that proper boundaries cannot be derived from mixed tests.\n\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d655490417ee22da3267fe6592a0ec2023c3c0db", "tree": "d6a46018c9e1e644757d89b763e09a4472254d16", "parents": [ "4cabc5b186b5427b9ee5a7495172542af105f02b" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Jul 21 00:00:22 2017 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:20:27 2017 -0700" }, "message": "bpf: allow to specify log level and reduce it for test_verifier\n\nFor the test_verifier case, it\u0027s quite hard to parse log level 2 to\nfigure out what\u0027s causing an issue when used to log level 1. We do\nwant to use bpf_verify_program() in order to simulate some of the\ntests with strict alignment. So just add an argument to pass the level\nand put it to 1 for test_verifier.\n\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4cabc5b186b5427b9ee5a7495172542af105f02b", "tree": "be016374d6f8a4d6dd85af1836af394ef2c50520", "parents": [ "6399f1fae4ec29fab5ec76070435555e256ca3a6" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Jul 21 00:00:21 2017 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Jul 20 15:20:27 2017 -0700" }, "message": "bpf: fix mixed signed/unsigned derived min/max value bounds\n\nEdward reported that there\u0027s an issue in min/max value bounds\ntracking when signed and unsigned compares both provide hints\non limits when having unknown variables. E.g. a program such\nas the following should have been rejected:\n\n 0: (7a) *(u64 *)(r10 -8) \u003d 0\n 1: (bf) r2 \u003d r10\n 2: (07) r2 +\u003d -8\n 3: (18) r1 \u003d 0xffff8a94cda93400\n 5: (85) call bpf_map_lookup_elem#1\n 6: (15) if r0 \u003d\u003d 0x0 goto pc+7\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R10\u003dfp\n 7: (7a) *(u64 *)(r10 -16) \u003d -8\n 8: (79) r1 \u003d *(u64 *)(r10 -16)\n 9: (b7) r2 \u003d -1\n 10: (2d) if r1 \u003e r2 goto pc+3\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R1\u003dinv,min_value\u003d0\n R2\u003dimm-1,max_value\u003d18446744073709551615,min_align\u003d1 R10\u003dfp\n 11: (65) if r1 s\u003e 0x1 goto pc+2\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R1\u003dinv,min_value\u003d0,max_value\u003d1\n R2\u003dimm-1,max_value\u003d18446744073709551615,min_align\u003d1 R10\u003dfp\n 12: (0f) r0 +\u003d r1\n 13: (72) *(u8 *)(r0 +0) \u003d 0\n R0\u003dmap_value_adj(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d1 R1\u003dinv,min_value\u003d0,max_value\u003d1\n R2\u003dimm-1,max_value\u003d18446744073709551615,min_align\u003d1 R10\u003dfp\n 14: (b7) r0 \u003d 0\n 15: (95) exit\n\nWhat happens is that in the first part ...\n\n 8: (79) r1 \u003d *(u64 *)(r10 -16)\n 9: (b7) r2 \u003d -1\n 10: (2d) if r1 \u003e r2 goto pc+3\n\n... r1 carries an unsigned value, and is compared as unsigned\nagainst a register carrying an immediate. Verifier deduces in\nreg_set_min_max() that since the compare is unsigned and operation\nis greater than (\u003e), that in the fall-through/false case, r1\u0027s\nminimum bound must be 0 and maximum bound must be r2. Latter is\nlarger than the bound and thus max value is reset back to being\n\u0027invalid\u0027 aka BPF_REGISTER_MAX_RANGE. Thus, r1 state is now\n\u0027R1\u003dinv,min_value\u003d0\u0027. The subsequent test ...\n\n 11: (65) if r1 s\u003e 0x1 goto pc+2\n\n... is a signed compare of r1 with immediate value 1. Here,\nverifier deduces in reg_set_min_max() that since the compare\nis signed this time and operation is greater than (\u003e), that\nin the fall-through/false case, we can deduce that r1\u0027s maximum\nbound must be 1, meaning with prior test, we result in r1 having\nthe following state: R1\u003dinv,min_value\u003d0,max_value\u003d1. Given that\nthe actual value this holds is -8, the bounds are wrongly deduced.\nWhen this is being added to r0 which holds the map_value(_adj)\ntype, then subsequent store access in above case will go through\ncheck_mem_access() which invokes check_map_access_adj(), that\nwill then probe whether the map memory is in bounds based\non the min_value and max_value as well as access size since\nthe actual unknown value is min_value \u003c\u003d x \u003c\u003d max_value; commit\nfce366a9dd0d (\"bpf, verifier: fix alu ops against map_value{,\n_adj} register types\") provides some more explanation on the\nsemantics.\n\nIt\u0027s worth to note in this context that in the current code,\nmin_value and max_value tracking are used for two things, i)\ndynamic map value access via check_map_access_adj() and since\ncommit 06c1c049721a (\"bpf: allow helpers access to variable memory\")\nii) also enforced at check_helper_mem_access() when passing a\nmemory address (pointer to packet, map value, stack) and length\npair to a helper and the length in this case is an unknown value\ndefining an access range through min_value/max_value in that\ncase. The min_value/max_value tracking is /not/ used in the\ndirect packet access case to track ranges. However, the issue\nalso affects case ii), for example, the following crafted program\nbased on the same principle must be rejected as well:\n\n 0: (b7) r2 \u003d 0\n 1: (bf) r3 \u003d r10\n 2: (07) r3 +\u003d -512\n 3: (7a) *(u64 *)(r10 -16) \u003d -8\n 4: (79) r4 \u003d *(u64 *)(r10 -16)\n 5: (b7) r6 \u003d -1\n 6: (2d) if r4 \u003e r6 goto pc+5\n R1\u003dctx R2\u003dimm0,min_value\u003d0,max_value\u003d0,min_align\u003d2147483648 R3\u003dfp-512\n R4\u003dinv,min_value\u003d0 R6\u003dimm-1,max_value\u003d18446744073709551615,min_align\u003d1 R10\u003dfp\n 7: (65) if r4 s\u003e 0x1 goto pc+4\n R1\u003dctx R2\u003dimm0,min_value\u003d0,max_value\u003d0,min_align\u003d2147483648 R3\u003dfp-512\n R4\u003dinv,min_value\u003d0,max_value\u003d1 R6\u003dimm-1,max_value\u003d18446744073709551615,min_align\u003d1\n R10\u003dfp\n 8: (07) r4 +\u003d 1\n 9: (b7) r5 \u003d 0\n 10: (6a) *(u16 *)(r10 -512) \u003d 0\n 11: (85) call bpf_skb_load_bytes#26\n 12: (b7) r0 \u003d 0\n 13: (95) exit\n\nMeaning, while we initialize the max_value stack slot that the\nverifier thinks we access in the [1,2] range, in reality we\npass -7 as length which is interpreted as u32 in the helper.\nThus, this issue is relevant also for the case of helper ranges.\nResetting both bounds in check_reg_overflow() in case only one\nof them exceeds limits is also not enough as similar test can be\ncreated that uses values which are within range, thus also here\nlearned min value in r1 is incorrect when mixed with later signed\ntest to create a range:\n\n 0: (7a) *(u64 *)(r10 -8) \u003d 0\n 1: (bf) r2 \u003d r10\n 2: (07) r2 +\u003d -8\n 3: (18) r1 \u003d 0xffff880ad081fa00\n 5: (85) call bpf_map_lookup_elem#1\n 6: (15) if r0 \u003d\u003d 0x0 goto pc+7\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R10\u003dfp\n 7: (7a) *(u64 *)(r10 -16) \u003d -8\n 8: (79) r1 \u003d *(u64 *)(r10 -16)\n 9: (b7) r2 \u003d 2\n 10: (3d) if r2 \u003e\u003d r1 goto pc+3\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R1\u003dinv,min_value\u003d3\n R2\u003dimm2,min_value\u003d2,max_value\u003d2,min_align\u003d2 R10\u003dfp\n 11: (65) if r1 s\u003e 0x4 goto pc+2\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0\n R1\u003dinv,min_value\u003d3,max_value\u003d4 R2\u003dimm2,min_value\u003d2,max_value\u003d2,min_align\u003d2 R10\u003dfp\n 12: (0f) r0 +\u003d r1\n 13: (72) *(u8 *)(r0 +0) \u003d 0\n R0\u003dmap_value_adj(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d3,max_value\u003d4\n R1\u003dinv,min_value\u003d3,max_value\u003d4 R2\u003dimm2,min_value\u003d2,max_value\u003d2,min_align\u003d2 R10\u003dfp\n 14: (b7) r0 \u003d 0\n 15: (95) exit\n\nThis leaves us with two options for fixing this: i) to invalidate\nall prior learned information once we switch signed context, ii)\nto track min/max signed and unsigned boundaries separately as\ndone in [0]. (Given latter introduces major changes throughout\nthe whole verifier, it\u0027s rather net-next material, thus this\npatch follows option i), meaning we can derive bounds either\nfrom only signed tests or only unsigned tests.) There is still the\ncase of adjust_reg_min_max_vals(), where we adjust bounds on ALU\noperations, meaning programs like the following where boundaries\non the reg get mixed in context later on when bounds are merged\non the dst reg must get rejected, too:\n\n 0: (7a) *(u64 *)(r10 -8) \u003d 0\n 1: (bf) r2 \u003d r10\n 2: (07) r2 +\u003d -8\n 3: (18) r1 \u003d 0xffff89b2bf87ce00\n 5: (85) call bpf_map_lookup_elem#1\n 6: (15) if r0 \u003d\u003d 0x0 goto pc+6\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R10\u003dfp\n 7: (7a) *(u64 *)(r10 -16) \u003d -8\n 8: (79) r1 \u003d *(u64 *)(r10 -16)\n 9: (b7) r2 \u003d 2\n 10: (3d) if r2 \u003e\u003d r1 goto pc+2\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R1\u003dinv,min_value\u003d3\n R2\u003dimm2,min_value\u003d2,max_value\u003d2,min_align\u003d2 R10\u003dfp\n 11: (b7) r7 \u003d 1\n 12: (65) if r7 s\u003e 0x0 goto pc+2\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R1\u003dinv,min_value\u003d3\n R2\u003dimm2,min_value\u003d2,max_value\u003d2,min_align\u003d2 R7\u003dimm1,max_value\u003d0 R10\u003dfp\n 13: (b7) r0 \u003d 0\n 14: (95) exit\n\n from 12 to 15: R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0\n R1\u003dinv,min_value\u003d3 R2\u003dimm2,min_value\u003d2,max_value\u003d2,min_align\u003d2 R7\u003dimm1,min_value\u003d1 R10\u003dfp\n 15: (0f) r7 +\u003d r1\n 16: (65) if r7 s\u003e 0x4 goto pc+2\n R0\u003dmap_value(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d0,max_value\u003d0 R1\u003dinv,min_value\u003d3\n R2\u003dimm2,min_value\u003d2,max_value\u003d2,min_align\u003d2 R7\u003dinv,min_value\u003d4,max_value\u003d4 R10\u003dfp\n 17: (0f) r0 +\u003d r7\n 18: (72) *(u8 *)(r0 +0) \u003d 0\n R0\u003dmap_value_adj(ks\u003d8,vs\u003d8,id\u003d0),min_value\u003d4,max_value\u003d4 R1\u003dinv,min_value\u003d3\n R2\u003dimm2,min_value\u003d2,max_value\u003d2,min_align\u003d2 R7\u003dinv,min_value\u003d4,max_value\u003d4 R10\u003dfp\n 19: (b7) r0 \u003d 0\n 20: (95) exit\n\nMeaning, in adjust_reg_min_max_vals() we must also reset range\nvalues on the dst when src/dst registers have mixed signed/\nunsigned derived min/max value bounds with one unbounded value\nas otherwise they can be added together deducing false boundaries.\nOnce both boundaries are established from either ALU ops or\ncompare operations w/o mixing signed/unsigned insns, then they\ncan safely be added to other regs also having both boundaries\nestablished. Adding regs with one unbounded side to a map value\nwhere the bounded side has been learned w/o mixing ops is\npossible, but the resulting map value won\u0027t recover from that,\nmeaning such op is considered invalid on the time of actual\naccess. Invalid bounds are set on the dst reg in case i) src reg,\nor ii) in case dst reg already had them. The only way to recover\nwould be to perform i) ALU ops but only \u0027add\u0027 is allowed on map\nvalue types or ii) comparisons, but these are disallowed on\npointers in case they span a range. This is fine as only BPF_JEQ\nand BPF_JNE may be performed on PTR_TO_MAP_VALUE_OR_NULL registers\nwhich potentially turn them into PTR_TO_MAP_VALUE type depending\non the branch, so only here min/max value cannot be invalidated\nfor them.\n\nIn terms of state pruning, value_from_signed is considered\nas well in states_equal() when dealing with adjusted map values.\nWith regards to breaking existing programs, there is a small\nrisk, but use-cases are rather quite narrow where this could\noccur and mixing compares probably unlikely.\n\nJoint work with Josef and Edward.\n\n [0] https://lists.iovisor.org/pipermail/iovisor-dev/2017-June/000822.html\n\nFixes: 484611357c19 (\"bpf: allow access into map value arrays\")\nReported-by: Edward Cree \u003cecree@solarflare.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nSigned-off-by: Edward Cree \u003cecree@solarflare.com\u003e\nSigned-off-by: Josef Bacik \u003cjbacik@fb.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "63a86362130f4c17eaa57f3ef5171ec43111a54e", "tree": "7255338325e2e8bf6ca17bf5eff357260ee53c62", "parents": [ "54a7d50b9205b5064628c1d10de6531d2d9fbc90", "ffa64d5e0d2f053d76ff9b1d2f437150b49c4112" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 14:56:46 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 14:56:46 2017 -0700" }, "message": "Merge tag \u0027pm-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management fixes from Rafael Wysocki:\n \"These are two stable-candidate fixes for the intel_pstate driver and\n the generic power domains (genpd) framework.\n\n Specifics:\n\n - Fix the average CPU load computations in the intel_pstate driver on\n Knights Landing (Xeon Phi) processors that require an extra factor\n to compensate for a rate change differences between the TSC and\n MPERF which is missing (Srinivas Pandruvada).\n\n - Fix an initialization ordering issue in the generic power domains\n (genpd) framework (Sudeep Holla)\"\n\n* tag \u0027pm-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n PM / Domains: defer dev_pm_domain_set() until genpd-\u003eattach_dev succeeds if present\n cpufreq: intel_pstate: Correct the busy calculation for KNL\n" }, { "commit": "54a7d50b9205b5064628c1d10de6531d2d9fbc90", "tree": "a51908ed455f210d1752a732c3af95c936bb6f63", "parents": [ "791f2df39b8248490ab41022939debaa82b96028" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 11:34:47 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 11:34:47 2017 -0700" }, "message": "x86: mark kprobe templates as character arrays, not single characters\n\nThey really are, and the \"take the address of a single character\" makes\nthe string fortification code unhappy (it believes that you can now only\nacccess one byte, rather than a byte range, and then raises errors for\nthe memory copies going on in there).\n\nWe could now remove a few \u0027addressof\u0027 operators (since arrays naturally\ndegrade to pointers), but this is the minimal patch that just changes\nthe C prototypes of those template arrays (the templates themselves are\ndefined in inline asm).\n\nReported-by: kernel test robot \u003cxiaolong.ye@intel.com\u003e\nAcked-and-tested-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\nCc: Daniel Micay \u003cdanielmicay@gmail.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "791f2df39b8248490ab41022939debaa82b96028", "tree": "99fd6e5536433d7758f087635e67e78a33749ae4", "parents": [ "465b0dbb38a02c449c62563ce0995e741c0ebe18", "84969465ddc4f8aeb3b993123b571aa01c5f2683" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:41:12 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:41:12 2017 -0700" }, "message": "Merge branch \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs\n\nPull misc filesystem fixes from Jan Kara:\n \"Several ACL related fixes for ext2, reiserfs, and hfsplus.\n\n And also one minor isofs cleanup\"\n\n* \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:\n hfsplus: Don\u0027t clear SGID when inheriting ACLs\n isofs: Fix off-by-one in \u0027session\u0027 mount option parsing\n reiserfs: preserve i_mode if __reiserfs_set_acl() fails\n ext2: preserve i_mode if ext2_set_acl() fails\n ext2: Don\u0027t clear SGID when inheriting ACLs\n reiserfs: Don\u0027t clear SGID when inheriting ACLs\n" }, { "commit": "465b0dbb38a02c449c62563ce0995e741c0ebe18", "tree": "83bc36f49488c2732de858555afdb49cd7897ff8", "parents": [ "f58781c983393716e3d57398eb980a31121a5a7f", "4db08d016ccedb5b97869724a096990acad59685" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:30:16 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:30:16 2017 -0700" }, "message": "Merge tag \u0027for-f2fs-v4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs\n\nPull f2fs fixes from Jaegeuk Kim:\n \"We\u0027ve filed some bug fixes:\n\n - missing f2fs case in terms of stale SGID bit, introduced by Jan\n\n - build error for seq_file.h\n\n - avoid cpu lockup\n\n - wrong inode_unlock in error case\"\n\n* tag \u0027for-f2fs-v4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs:\n f2fs: avoid cpu lockup\n f2fs: include seq_file.h for sysfs.c\n f2fs: Don\u0027t clear SGID when inheriting ACLs\n f2fs: remove extra inode_unlock() in error path\n" }, { "commit": "f58781c983393716e3d57398eb980a31121a5a7f", "tree": "c73a8d98da2844f5571221284478575026b23563", "parents": [ "ae1c908517b4d9e156f0eaed8b98238587c94250", "b0659ae5e30074ede1dc08f2c6d64f0c11d64e0f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:22:26 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:22:26 2017 -0700" }, "message": "Merge branch \u0027stable-4.13\u0027 of git://git.infradead.org/users/pcmoore/audit\n\nPull audit fix from Paul Moore:\n \"A small audit fix, just a single line, to plug a memory leak in some\n audit error handling code\"\n\n* \u0027stable-4.13\u0027 of git://git.infradead.org/users/pcmoore/audit:\n audit: fix memleak in auditd_send_unicast_skb.\n" }, { "commit": "ae1c908517b4d9e156f0eaed8b98238587c94250", "tree": "b127a89ac587758ee6ecf904dc999dd66bf7d1e8", "parents": [ "c6efb4548d0a665b791b3db0de7accd63d630f1d", "bbb3be170ac2891526ad07b18af7db226879a8e7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:17:53 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:17:53 2017 -0700" }, "message": "Merge tag \u0027libnvdimm-fixes-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm\n\nPull libnvdimm fixes from Dan Williams:\n \"A handful of small fixes for 4.13-rc2. Three of these fixes are tagged\n for -stable. They have all appeared in at least one -next release with\n no reported issues\n\n - Fix handling of media errors that span a sector\n\n - Fix support of multiple namespaces in a libnvdimm region being in\n device-dax mode\n\n - Clean up the machine check notifier properly when the nfit driver\n fails to register\n\n - Address a static analysis (smatch) report in device-dax\"\n\n* tag \u0027libnvdimm-fixes-4.13-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:\n device-dax: fix sysfs duplicate warnings\n MAINTAINERS: list drivers/acpi/nfit/ files for libnvdimm sub-system\n acpi/nfit: Fix memory corruption/Unregister mce decoder on failure\n device-dax: fix \u0027passing zero to ERR_PTR()\u0027 warning\n libnvdimm: fix badblock range handling of ARS range\n" }, { "commit": "c6efb4548d0a665b791b3db0de7accd63d630f1d", "tree": "50aaa2c8ead4d7a4e04d9b6d4b3c5738d11b5df3", "parents": [ "beaec533fc2701a28a4d667f67c9f59c6e4e0d13", "929b60a85b731d803f2b7829149496479eba0b20" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:14:54 2017 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 20 10:14:54 2017 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid\n\nPull HID fixes from Jiri Kosina:\n\n - HID multitouch 4.12 regression fix from Dmitry Torokhov\n\n - error handling fix for HID++ driver from Gustavo A. R. Silva\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid:\n HID: hid-logitech-hidpp: add NULL check on devm_kmemdup() return value\n HID: multitouch: do not blindly set EV_KEY or EV_ABS bits\n" }, { "commit": "ffa64d5e0d2f053d76ff9b1d2f437150b49c4112", "tree": "08a400aeadf632bceae9613658dacb8ac6b275bc", "parents": [ "5771a8c08880cdca3bfb4a3fc6d309d6bba20877", "6e34e1f23d780978da65968327cbba6d7013a73f", "975e83cfb8dc16e7a2fdc58188c77c0c605876c2" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Thu Jul 20 18:57:15 2017 +0200" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Thu Jul 20 18:57:15 2017 +0200" }, "message": "Merge branches \u0027intel_pstate\u0027 and \u0027pm-domains\u0027\n\n* intel_pstate:\n cpufreq: intel_pstate: Correct the busy calculation for KNL\n\n* pm-domains:\n PM / Domains: defer dev_pm_domain_set() until genpd-\u003eattach_dev succeeds if present\n" }, { "commit": "a62ab66b13a0f9bcb17b7b761f6670941ed5cd62", "tree": "0f14609ba1f3e3c95a7192d7f59aae349e728658", "parents": [ "5a7a88f1b488e4ee49eb3d5b82612d4d9ffdf2c3" ], "author": { "name": "Ismail, Mustafa", "email": "mustafa.ismail@intel.com", "time": "Fri Jul 14 09:41:31 2017 -0500" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:24:13 2017 -0400" }, "message": "RDMA/core: Initialize port_num in qp_attr\n\nInitialize the port_num for iWARP in rdma_init_qp_attr.\n\nFixes: 5ecce4c9b17b(\"Check port number supplied by user verbs cmds\")\nCc: \u003cstable@vger.kernel.org\u003e # v2.6.14+\nReviewed-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Mustafa Ismail \u003cmustafa.ismail@intel.com\u003e\nTested-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "5a7a88f1b488e4ee49eb3d5b82612d4d9ffdf2c3", "tree": "d6af1ba339e228d7b3fe40c1b7555f43795b3577", "parents": [ "963916fdb3e5ad4af57ac959b5a03bf23f7568ca" ], "author": { "name": "Ismail, Mustafa", "email": "mustafa.ismail@intel.com", "time": "Fri Jul 14 09:41:30 2017 -0500" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:24:13 2017 -0400" }, "message": "RDMA/uverbs: Fix the check for port number\n\nThe port number is only valid if IB_QP_PORT is set in the mask.\nSo only check port number if it is valid to prevent modify_qp from\nfailing due to an invalid port number.\n\nFixes: 5ecce4c9b17b(\"Check port number supplied by user verbs cmds\")\nCc: \u003cstable@vger.kernel.org\u003e # v2.6.14+\nReviewed-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Mustafa Ismail \u003cmustafa.ismail@intel.com\u003e\nTested-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "963916fdb3e5ad4af57ac959b5a03bf23f7568ca", "tree": "197b61b6b73cb5e661221f6307da2058867bc111", "parents": [ "e6e52aec494900912fedd7b595b8827ba70a670d" ], "author": { "name": "Kalderon, Michal", "email": "Michal.Kalderon@cavium.com", "time": "Thu Jul 06 23:22:11 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:24:13 2017 -0400" }, "message": "IB/cma: Fix reference count leak when no ipv4 addresses are set\n\nOnce in_dev_get is called to receive in_device pointer, the\nin_device reference counter is increased, but if there are\nno ipv4 addresses configured on the net-device the ifa_list\nwill be null, resulting in a flow that doesn\u0027t call in_dev_put\nto decrease the ref_cnt.\nThis was exposed when running RoCE over ipv6 without any ipv4\naddresses configured\n\nFixes: commit 8e3867310c90 (\"IB/cma: Fix a race condition in iboe_addr_get_sgid()\")\n\nSigned-off-by: Michal Kalderon \u003cMichal.Kalderon@cavium.com\u003e\nSigned-off-by: Ariel Elior \u003cAriel.Elior@cavium.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "e6e52aec494900912fedd7b595b8827ba70a670d", "tree": "8efe6cb362d38cf717cd79ad583653e917997cde", "parents": [ "1217197142d1681a8b8aaa88cdf4b245b76974cd" ], "author": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Thu Jul 06 10:21:36 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/iser: don\u0027t send an rkey if all data is written as immadiate-data\n\nWe might get some bogus error completions in case the target will\nremotely invalidate the rkey and the HCA will need to retransmit\nfrom this buffer.\n\nSigned-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "1217197142d1681a8b8aaa88cdf4b245b76974cd", "tree": "54b814d43b57f593b3a6946e10495b6e382b5d05", "parents": [ "c75d3ec8c0ee469de79ae83c1a827d753603e49f" ], "author": { "name": "Vijay Immanuel", "email": "vijayi@attalasystems.com", "time": "Tue Jun 27 12:19:38 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "rxe: fix broken receive queue draining\n\nIf we modified the qp to ERROR state, and\ndrained the recieve queue, post_recv must\ntrigger the responder task to complete\nthe drain work request.\n\nCc: Bart Van Assche \u003cbart.vanassche@sandisk.com\u003e\nSigned-off-by: Vijay Immanuel \u003cvijayi@attalasystems.com\u003e\nSigned-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\nReviewed-by: Bart Van Assche \u003cbart.vanassche@wdc.com\u003e--\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "c75d3ec8c0ee469de79ae83c1a827d753603e49f", "tree": "49ac00312ed0d4af72a8254e95baa98450f3c089", "parents": [ "720336c42e41a917002fcae3aa14e30f5022bbb7" ], "author": { "name": "Amrani, Ram", "email": "Ram.Amrani@cavium.com", "time": "Mon Jun 26 19:05:04 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/qedr: Prevent memory overrun in verbs\u0027 user responses\n\nWrap ib_copy_to_udata with a function that ensures that the data\nbeing copied over to user space isn\u0027t longer than the allowed.\n\nFixes: cecbcddf6461 (\"qedr: Add support for QP verbs\")\nFixes: a7efd7773e31 (\"qedr: Add support for PD,PKEY and CQ verbs\")\nFixes: ac1b36e55a51 (\"qedr: Add support for user context verbs\")\nSigned-off-by: Ram Amrani \u003cRam.Amrani@cavium.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "720336c42e41a917002fcae3aa14e30f5022bbb7", "tree": "727fa0c4c57aef84be2f41a757893dc9ed4e5c39", "parents": [ "4542e3c79a2c5a167cbeb4f4190d5f705d272002" ], "author": { "name": "Ganesh Goudar", "email": "ganeshgr@chelsio.com", "time": "Wed Jun 21 19:55:43 2017 +0530" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "iw_cxgb4: don\u0027t use WR keys/addrs for 0 byte reads\n\nOnly use the read sge lkey/addr and the remote rkey/addr if the\nlength of the read is not zero. Otherwise the read response might\nbe treated as the RTR read response and not delivered to the\napplication. Or worse Terminator hardware will fail a 0B read\nif the STAG is 0 even if the read length is 0.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Ganesh Goudar \u003cganeshgr@chelsio.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "4542e3c79a2c5a167cbeb4f4190d5f705d272002", "tree": "13fc15f92dad0a2307f26ba91e7f4721d0e3ee06", "parents": [ "a25ce4270bfdd522207b02f81a594c7d1746b697" ], "author": { "name": "Håkon Bugge", "email": "haakon.bugge@oracle.com", "time": "Tue Jun 20 14:07:50 2017 +0200" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "IB/mlx4: Fix CM REQ retries in paravirt mode\n\nCM REQs cannot be successfully retried, because a new pv_cm_id is\ncreated for each request, without checking if one already exists.\n\nBy checking if an id exists before creating one, the bug is fixed.\n\nThis bug can be provoked by running an RDMA CM user-land application,\nbut inserting a five seconds delay before the rdma_accept() call on\nthe passive side. This delay is larger than the default CMA timeout,\nand triggers a retry from the active side. The retried REQ will use\nanother pv_cm_id (the cm_id on the wire). This confuses the CM\nprotocol and two REJs are sent from the passive side.\n\nHere is an excerpt from ibdump running without the patch:\n\n3.285092 LID: 4 -\u003e LID: 4 SDP 290 CM: ConnectRequest(SDP Hello)\n7.382711 LID: 4 -\u003e LID: 4 SDP 290 CM: ConnectRequest(SDP Hello)\n7.382861 LID: 4 -\u003e LID: 4 InfiniBand 290 CM: ConnectReject\n7.387644 LID: 4 -\u003e LID: 4 InfiniBand 290 CM: ConnectReject\n\nand here is the same with bug fix applied:\n\n3.251010 LID: 4 -\u003e LID: 4 SDP 290 CM: ConnectRequest(SDP Hello)\n7.349387 LID: 4 -\u003e LID: 4 SDP 290 CM: ConnectRequest(SDP Hello)\n8.258443 LID: 4 -\u003e LID: 4 SDP 290 CM: ConnectReply(SDP Hello)\n8.259890 LID: 4 -\u003e LID: 4 InfiniBand 290 CM: ReadyToUse\n\nSuggested-by: Venkat Venkatsubra \u003cvenkat.x.venkatsubra@oracle.com\u003e\nSigned-off-by: Håkon Bugge \u003chaakon.bugge@oracle.com\u003e\nReported-by: Wei Lin Guay \u003cwei.lin.guay@oracle.com\u003e\nTested-by: Wei Lin Guay \u003cwei.lin.guay@oracle.com\u003e\nReviewed-by: Yuval Shaia \u003cyuval.shaia@oracle.com\u003e\nAcked-by: Jack Morgenstein \u003cjackm@dev.mellanox.co.il\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "a25ce4270bfdd522207b02f81a594c7d1746b697", "tree": "1abc324af4c736e444135f1c6aa0f97201611048", "parents": [ "266098b841d48f7f0db40424bdbc072e4db14e9b" ], "author": { "name": "Kaike Wan", "email": "kaike.wan@intel.com", "time": "Sat Jun 17 10:37:26 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "IB/rdmavt: Setting of QP timeout can overflow jiffies computation\n\nCurrent computation of qp-\u003etimeout_jiffies in rvt_modify_qp() will cause\noverflow due to the fact that the input to the function usecs_to_jiffies\nis only 32-bit ( unsigned int). Overflow will occur when attr-\u003etimeout is\nequal to or greater than 30. The consequence is unnecessarily excessive\nretry and thus degradation of the system performance.\n\nThis patch fixes the problem by limiting the input to 5-bit and calling\nusecs_to_jiffies() before multiplying the scaling factor.\n\nReviewed-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nSigned-off-by: Kaike Wan \u003ckaike.wan@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "266098b841d48f7f0db40424bdbc072e4db14e9b", "tree": "8780f3e68acf33605c46683c2ade012f25d715a3", "parents": [ "601577b7d14cba5848b4c44ef719881f8538f702" ], "author": { "name": "Matan Barak", "email": "matanb@mellanox.com", "time": "Thu Jun 08 17:30:47 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "IB/core: Fix sparse warnings\n\nDelete unused variables to prevent sparse warnings.\n\nFixes: db1b5ddd5336 (\"IB/core: Rename uverbs event file structure\")\nFixes: fd3c7904db6e (\"IB/core: Change idr objects to use the new schema\")\n\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "601577b7d14cba5848b4c44ef719881f8538f702", "tree": "3e97439b73b2055fd6ead7f2d5dac70544640fe5", "parents": [ "499e456981d376ef614e257d5dcb280581db9ac6" ], "author": { "name": "Selvin Xavier", "email": "selvin.xavier@broadcom.com", "time": "Thu Jun 29 12:28:19 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/bnxt_re: Fix the value reported for local ack delay\n\nLocal ack delay exposed by the driver is 0 which means infinite QP\ntimeout. Reporting the default value to 16 (approx 260ms)\n\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "499e456981d376ef614e257d5dcb280581db9ac6", "tree": "cc23bc5c9e008a67b134f591f7930836a069c4b9", "parents": [ "10d1dedf9b428ed776f244eb8b43f28974da3fb0" ], "author": { "name": "Selvin Xavier", "email": "selvin.xavier@broadcom.com", "time": "Thu Jun 29 12:28:18 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/bnxt_re: Report MISSED_EVENTS in req_notify_cq\n\nWhile invoking the req_notify_cq hook, ULPs can request\nwhether the CQs have any CQEs pending. If CQEs are pending,\ndrivers can indicate it by returning 1 for req_notify_cq.\nThe stack will poll CQ again till CQ is empty.\n\nThis patch peeks the CQ for any valid entries and return accordingly.\n\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "10d1dedf9b428ed776f244eb8b43f28974da3fb0", "tree": "a22252253c2a494546587640fb30700fbbd987d4", "parents": [ "254cd2590d408f2c0375fbf73ef8250d47529e7b" ], "author": { "name": "Devesh Sharma", "email": "devesh.sharma@broadcom.com", "time": "Thu Jun 29 12:28:17 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/bnxt_re: Fix return value of poll routine\n\nFix the incorrect reporting of number of polled\nentries by taking into account the max CQ depth\nin the driver.\n\nSigned-off-by: Devesh Sharma \u003cdevesh.sharma@broadcom.com\u003e\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nReviewed-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "254cd2590d408f2c0375fbf73ef8250d47529e7b", "tree": "48f0a0954f92d87fd841f71f122dd486537d0c1f", "parents": [ "536f092805cecc5c2c0ba7f051a7552619bd4491" ], "author": { "name": "Devesh Sharma", "email": "devesh.sharma@broadcom.com", "time": "Thu Jun 29 12:28:16 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/bnxt_re: Enable atomics only if host bios supports\n\nDriver shall check if the host system bios has enabled\nAtomic operations capability in PCI Device Control 2\nregister of the pci-device. Expose the ATOMIC_HCA\nflag only if the Atomic operations capability is set.\n\nSigned-off-by: Devesh Sharma \u003cdevesh.sharma@broadcom.com\u003e\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "536f092805cecc5c2c0ba7f051a7552619bd4491", "tree": "814afb5d751ce41bfaaf9c53486c24110b5d4c2a", "parents": [ "a25d112fe9c8e8817cde1df17a82aee472c55993" ], "author": { "name": "Somnath Kotur", "email": "somnath.kotur@broadcom.com", "time": "Thu Jun 29 12:28:14 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/bnxt_re: Specify RDMA component when allocating stats context\n\nStarting FW version 20.6.47, firmware is keeping separate statistics\nfor L2 and RDMA. However, driver needs to specify RDMA or not when\nallocating stat_ctx.\n\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "a25d112fe9c8e8817cde1df17a82aee472c55993", "tree": "644616f02fe039323fa385d82835b213bb4e42a4", "parents": [ "58d4a671d0eac45db1c7f27c8684c277249ac127" ], "author": { "name": "Eddie Wai", "email": "eddie.wai@broadcom.com", "time": "Thu Jun 29 12:28:13 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/bnxt_re: Fixed the max_rd_atomic support for initiator and destination QP\n\nThere\u0027s a couple of bugs in the support of max_rd_atomic and\nmax_dest_rd_atomic. In the modify_qp, if the requested max_rd_atomic,\nwhich is the ORRQ size, is greater than what the chip can support,\nthen we have to cap the request to chip max as we can\u0027t have the HW\noverflow the ORRQ. Capping the max_rd_atomic support internally is okay\nto do as the remaining read/atomic WRs will still be sitting in the SQ.\nHowever, for the max_dest_rd_atomic, the driver has to error out as\nthis dictates the IRRQ size and we can\u0027t control what the remote\nside sends.\n\nSigned-off-by: Eddie Wai \u003ceddie.wai@broadcom.com\u003e\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "58d4a671d0eac45db1c7f27c8684c277249ac127", "tree": "17b9b5225ac83a4120a4fb5c70040ebe1bb94a95", "parents": [ "4a62c5e9e2e1c15ceb1654715d9284d97f921119" ], "author": { "name": "Selvin Xavier", "email": "selvin.xavier@broadcom.com", "time": "Thu Jun 29 12:28:12 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/bnxt_re: Report supported value to IB stack in query_device\n\n- Report supported value for max_mr_size to IB stack in query_device.\n Also, check and log if MR size requested by application in\n reg_user_mr() is greater than value currently supported by driver.\n - Report only 4K page size support for now\n - Fix Max_QP value returned by ibv_devinfo -vv.\n In case of PF, FW reserves 129 QPs for creating QP1s of VFs\n and PF. So the max_qp value reported by FW for PF doesn\u0027tt include\n the QP1. Fixing this issue by adding 1 with the value reported\n by FW.\n\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "4a62c5e9e2e1c15ceb1654715d9284d97f921119", "tree": "ddb0c7c04a3170a048d54458007ff65a6a7cbeda", "parents": [ "ab69d4c8da38024191f3514c1296e9e8deea4e98" ], "author": { "name": "Selvin Xavier", "email": "selvin.xavier@broadcom.com", "time": "Thu Jun 29 12:28:11 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:50 2017 -0400" }, "message": "RDMA/bnxt_re: Do not free the ctx_tbl entry if delete GID fails\n\nThis fix is added only to avoid system crash in some a\nspecific scenario. When bnxt_re driver is loaded and if\nuser tries to change interface mac address, delete GID\nfails because QP1 is still associated with existing MAC\n(default GID). If the above command fails GID tables are\nnot modified in the h/w or driver, but the GID context memory\nis freed. Now, if the user changes the mac back to the original\nvalue, another add_gid comes to the driver where the driver\nreports that the GID is already present in its table\nand tries to access the context which was already freed.\n\nSo, in this case, in order to avoid NULL pointer de-reference,\nthis patch removes the context memory free if delete_gid fails\nand the same context memory is re-used in new add_gid.\nMemory cleanup will be taken care during driver unload, while\ndeleting the GID table.\n\nSigned-off-by: Kalesh AP \u003ckalesh-anakkur.purayil@broadcom.com\u003e\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "ab69d4c8da38024191f3514c1296e9e8deea4e98", "tree": "618d110aa7b9ca30ea560259a49d127d817bfdac", "parents": [ "b3b2c7c5506bd8d15214136f8b11a2e6c9728033" ], "author": { "name": "Somnath Kotur", "email": "somnath.kotur@broadcom.com", "time": "Thu Jun 29 12:28:09 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:49 2017 -0400" }, "message": "RDMA/bnxt_re: Fix WQE Size posted to HW to prevent it from throwing error\n\nPosting WQE size of 2 results in a WQE_FORMAT_ERROR\nthrown by the HW as it requires host to supply WQE Size with room\nfor atleast one SGE so that the resulting WQE size be atleast 3.\n\nSigned-off-by: Somnath Kotur \u003csomnath.kotur@broadcom.com\u003e\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "b3b2c7c5506bd8d15214136f8b11a2e6c9728033", "tree": "7217e72d2ffd31c887ff1395f7f97db3d99e20d2", "parents": [ "396551eb00e46aa8f843c448bced0c76971ec58c" ], "author": { "name": "Devesh Sharma", "email": "devesh.sharma@broadcom.com", "time": "Thu Jun 29 12:28:08 2017 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:49 2017 -0400" }, "message": "RDMA/bnxt_re: Free doorbell page index (DPI) during dealloc ucontext\n\nThe driver must free the DPI during the dealloc_ucontext\ninstead of freeing it during dealloc_pd. However, the DPI\nallocation scheme remains unchanged.\n\nSigned-off-by: Devesh Sharma \u003cdevesh.sharma@broadcom.com\u003e\nSigned-off-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "396551eb00e46aa8f843c448bced0c76971ec58c", "tree": "665ae6b298c56fded8fb25e0bc78f2a968de8537", "parents": [ "f0c6e88288d65c93bbc7da4fb6f7d51b2733228a" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Wed Jun 14 13:20:09 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:49 2017 -0400" }, "message": "IB/mlx5: Fix a warning message\n\n\"umem\" is a valid pointer. We intended to print \"*umem\" or even just\n\"err\" instead.\n\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nAcked-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "f0c6e88288d65c93bbc7da4fb6f7d51b2733228a", "tree": "f6881443a03d9731afaa8f78d69c5241dd7d76f3", "parents": [ "dd75cfa6d3216c79c695f5af13e52208afe374ad" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Thu Jul 13 10:46:49 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:49 2017 -0400" }, "message": "RDMA/ocrdma: Fix error codes in ocrdma_create_srq()\n\nIf either of these allocations fail then we return ERR_PTR(0). That\u0027s\nequivalent to NULL and results in a NULL pointer dereference in the\ncaller.\n\nFixes: fe2caefcdf58 (\"RDMA/ocrdma: Add driver for Emulex OneConnect IBoE RDMA adapter\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "dd75cfa6d3216c79c695f5af13e52208afe374ad", "tree": "1c00d9c1cf14071a97120ec7a5538ac551ba1cb3", "parents": [ "9064d6055c14f700aa13f7c72fd3e63d12bee643" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Thu Jul 13 10:46:14 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:49 2017 -0400" }, "message": "RDMA/ocrdma: Fix an error code in ocrdma_alloc_pd()\n\nWe should preserve the original \"status\" error code instead of resetting\nit to zero. Returning ERR_PTR(0) is the same as NULL and results in a\nNULL dereference in the callers. I added a printk() on error instead.\n\nFixes: 45e86b33ec8b (\"RDMA/ocrdma: Cache recv DB until QP moved to RTR\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "9064d6055c14f700aa13f7c72fd3e63d12bee643", "tree": "b638689f886be6aa1cd4652f9f0d5c13f3c007f0", "parents": [ "6ebedacbb44602d4dec3348dee5ec31dd9b09521" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Thu Jul 13 10:48:00 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:49 2017 -0400" }, "message": "IB/cxgb3: Fix error codes in iwch_alloc_mr()\n\nWe accidentally don\u0027t set the error code on some error paths. It means\nreturn ERR_PTR(0) which is NULL and results in a NULL dereference in the\ncaller.\n\nFixes: 13a239330abd (\"RDMA/cxgb3: Don\u0027t ignore insert_handle() failures\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nAcked-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "6ebedacbb44602d4dec3348dee5ec31dd9b09521", "tree": "da8292c509436a3867c9a711869afc32957ad28a", "parents": [ "6031e079aa4656743298ea235b894ee883f45c71" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Thu Jul 13 10:47:40 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:49 2017 -0400" }, "message": "cxgb4: Fix error codes in c4iw_create_cq()\n\nIf one of these kmalloc() calls fails then we return ERR_PTR(0) which is\nNULL. It results in a NULL dereference in the callers.\n\nFixes: cfdda9d76436 (\"RDMA/cxgb4: Add driver for Chelsio T4 RNIC\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nAcked-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "6031e079aa4656743298ea235b894ee883f45c71", "tree": "23599baf40234cbd2fdbbd4818315cd499e42ed8", "parents": [ "5c8857b653e71a9850a02837e1268e3198abbd1a" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Thu Jul 13 10:47:22 2017 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Thu Jul 20 11:20:49 2017 -0400" }, "message": "IB/i40iw: Fix error code in i40iw_create_cq()\n\nWe accidentally forgot to set the error code if ib_copy_from_udata()\nfails. It means we return ERR_PTR(0) which is NULL and results in a\nNULL dereference in the callers.\n\nFixes: d37498417947 (\"i40iw: add files for iwarp interface\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nAcked-by: Shiraz Saleem \u003cshiraz.saleem@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" } ], "next": "5c8857b653e71a9850a02837e1268e3198abbd1a" }