{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":643591137,"defaultBranch":"lineage-21.0","name":"android_kernel_htc_msm8998","ownerLogin":"TARKZiM","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2023-05-21T16:42:13.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/6601824?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1725721647.0","currentOid":""},"activityList":{"items":[{"before":"6e44c5fe2bbdf22924c0a8c9c5fc8e0bee7e368a","after":"78eb588dc5e5942c1145adec6e89c47d5b0ce8af","ref":"refs/heads/lineage-20.0","pushedAt":"2024-09-18T18:19:44.000Z","pushType":"push","commitsCount":9,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"UPSTREAM: loop: drop caches if offset or block_size are changed\n\nIf we don't drop caches used in old offset or block_size, we can get old data\nfrom new offset/block_size, which gives unexpected data to user.\n\nFor example, Martijn found a loopback bug in the below scenario.\n1) LOOP_SET_FD loads first two pages on loop file\n2) LOOP_SET_STATUS64 changes the offset on the loop file\n3) mount is failed due to the cached pages having wrong superblock\n\nCc: Jens Axboe \nCc: linux-block@vger.kernel.org\nReported-by: Martijn Coenen \nReviewed-by: Bart Van Assche \nSigned-off-by: Jaegeuk Kim \nSigned-off-by: Jens Axboe \n(cherry picked from commit 490b8c65b9db45896769e1095e78725775f47b3e)\nChange-Id: Iffb7e1f04ab587e1a8785bc862a425efb654be24\nSigned-off-by: JAISHNAVPRASAD ","shortMessageHtmlLink":"UPSTREAM: loop: drop caches if offset or block_size are changed"}},{"before":null,"after":"280bd30fd895fabff4f3a8cb89067bb7241397e5","ref":"refs/heads/lineage-21.0-ocm","pushedAt":"2024-09-07T15:07:27.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"defconfig: Initial OCM configs based on OCN","shortMessageHtmlLink":"defconfig: Initial OCM configs based on OCN"}},{"before":"a4e17fc703d71a667e7390a145860b70fc4aa97e","after":"476859e5522474685c031a3450fabd90fe03c715","ref":"refs/heads/lineage-21.0","pushedAt":"2024-09-06T17:56:43.000Z","pushType":"push","commitsCount":9,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"UPSTREAM: loop: drop caches if offset or block_size are changed\n\nIf we don't drop caches used in old offset or block_size, we can get old data\nfrom new offset/block_size, which gives unexpected data to user.\n\nFor example, Martijn found a loopback bug in the below scenario.\n1) LOOP_SET_FD loads first two pages on loop file\n2) LOOP_SET_STATUS64 changes the offset on the loop file\n3) mount is failed due to the cached pages having wrong superblock\n\nCc: Jens Axboe \nCc: linux-block@vger.kernel.org\nReported-by: Martijn Coenen \nReviewed-by: Bart Van Assche \nSigned-off-by: Jaegeuk Kim \nSigned-off-by: Jens Axboe \n(cherry picked from commit 490b8c65b9db45896769e1095e78725775f47b3e)\nChange-Id: Iffb7e1f04ab587e1a8785bc862a425efb654be24\nSigned-off-by: JAISHNAVPRASAD ","shortMessageHtmlLink":"UPSTREAM: loop: drop caches if offset or block_size are changed"}},{"before":"69e10cea2543217f1f21edddfa7c085d26229e9e","after":"a4e17fc703d71a667e7390a145860b70fc4aa97e","ref":"refs/heads/lineage-21.0","pushedAt":"2024-02-10T06:28:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"Revert \"arm64: dts: msm8998-ocn: Don't mount system partition early\"\n\nThis reverts commit 02862a4337b8c3d371c059f37a1a4122cfc0e755.","shortMessageHtmlLink":"Revert \"arm64: dts: msm8998-ocn: Don't mount system partition early\""}},{"before":"28fb2c45d09d80cbf321dbe86d444a6bd1cac4c0","after":"69e10cea2543217f1f21edddfa7c085d26229e9e","ref":"refs/heads/lineage-21.0","pushedAt":"2023-11-28T05:24:44.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"include: Fix bpf build errors","shortMessageHtmlLink":"include: Fix bpf build errors"}},{"before":"02862a4337b8c3d371c059f37a1a4122cfc0e755","after":"28fb2c45d09d80cbf321dbe86d444a6bd1cac4c0","ref":"refs/heads/lineage-21.0","pushedAt":"2023-11-28T04:01:09.000Z","pushType":"push","commitsCount":468,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"libfs: support RENAME_NOREPLACE in simple_rename()\n\nThis is trivial to do:\n\n - add flags argument to simple_rename()\n - check if flags doesn't have any other than RENAME_NOREPLACE\n - assign simple_rename() to .rename2 instead of .rename\n\nFilesystems converted:\n\nhugetlbfs, ramfs, bpf.\n\nDebugfs uses simple_rename() to implement debugfs_rename(), which is for\ndebugfs instances to rename files internally, not for userspace filesystem\naccess. For this case pass zero flags to simple_rename().\n\nChange-Id: I1a46ece3b40b05c9f18fd13b98062d2a959b76a0\nSigned-off-by: Miklos Szeredi \nAcked-by: Greg Kroah-Hartman \nCc: Alexei Starovoitov ","shortMessageHtmlLink":"libfs: support RENAME_NOREPLACE in simple_rename()"}},{"before":null,"after":"02862a4337b8c3d371c059f37a1a4122cfc0e755","ref":"refs/heads/lineage-21.0","pushedAt":"2023-11-27T18:00:59.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"arm64: dts: msm8998-ocn: Don't mount system partition early","shortMessageHtmlLink":"arm64: dts: msm8998-ocn: Don't mount system partition early"}},{"before":"7619ab176d8680afb739fd40dd0b4d3fc2b78def","after":"6e44c5fe2bbdf22924c0a8c9c5fc8e0bee7e368a","ref":"refs/heads/lineage-20.0","pushedAt":"2023-05-23T19:10:31.986Z","pushType":"push","commitsCount":2,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"BACKPORT: usb: gadget: rndis: prevent integer overflow in rndis_set_response()\n\ncommit 65f3324f4b6fed78b8761c3b74615ecf0ffa81fa upstream.\n\nIf \"BufOffset\" is very large the \"BufOffset + 8\" operation can have an\ninteger overflow.\n\nBug: 239842288\nCc: stable@kernel.org\nFixes: 38ea1eac7d88 (\"usb: gadget: rndis: check size of RNDIS_MSG_SET command\")\nSigned-off-by: Dan Carpenter \nLink: https://lore.kernel.org/r/20220301080424.GA17208@kili\nSigned-off-by: Greg Kroah-Hartman \nSigned-off-by: Lee Jones \nChange-Id: I4dcecb9ada2680a4211e5ccc9b27de2df964e404","shortMessageHtmlLink":"BACKPORT: usb: gadget: rndis: prevent integer overflow in rndis_set_r…"}},{"before":"be5893cb1a31b01eab816a9899ef82a05175f8b8","after":"7619ab176d8680afb739fd40dd0b4d3fc2b78def","ref":"refs/heads/lineage-20.0","pushedAt":"2023-05-22T19:11:20.312Z","pushType":"push","commitsCount":4,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"ANDROID: usb: f_accessory: Check buffer size when initialised via composite\n\nWhen communicating with accessory devices via USBFS, the initialisation\ncall-stack looks like:\n\n ConfigFS > Gadget ConfigFS > UDC > Gadget ConfigFS > Composite\n\nEventually ending up in composite_dev_prepare() where memory for the\ndata buffer is allocated and initialised. The default size used for the\nallocation is USB_COMP_EP0_BUFSIZ (4k). When handling bulk transfers,\nacc_ctrlrequest() needs to be able to handle buffers up to\nBULK_BUFFER_SIZE (16k). Instead of adding new generic attributes to\n'struct usb_request' to track the size of the allocated buffer, we can\nsimply split off the affected thread of execution to travel via a\nknowledgeable abstracted function acc_ctrlrequest_composite() where we\ncan complete the necessary specific checks.\n\nBug: 264029575\nSigned-off-by: Lee Jones \nChange-Id: Ia1280f85499621d3fa57f7262b4a2c80f4be7773\nSigned-off-by: Lee Jones ","shortMessageHtmlLink":"ANDROID: usb: f_accessory: Check buffer size when initialised via com…"}},{"before":null,"after":"be5893cb1a31b01eab816a9899ef82a05175f8b8","ref":"refs/heads/lineage-20.0","pushedAt":"2023-05-22T16:57:05.493Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"kbuild: use HOSTLDFLAGS for single .c executables\n\nWhen compiling executables from a single .c file, the linker is also\ninvoked. Pass the HOSTLDFLAGS like for other linker commands.\n\nSigned-off-by: Robin Jarry \nCc: Josh Poimboeuf \nCc: Ingo Molnar \nSigned-off-by: Masahiro Yamada \nChange-Id: I8332132775e7cc220eb7d1e20911223bca65e1e2","shortMessageHtmlLink":"kbuild: use HOSTLDFLAGS for single .c executables"}},{"before":null,"after":"76665b2f4e86e770eeac25afd466af482021556d","ref":"refs/heads/lineage-18.1","pushedAt":"2023-05-21T17:04:03.710Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"tty: Fix ->pgrp locking in tiocspgrp()\n\ncommit 54ffccbf053b5b6ca4f6e45094b942fab92a25fc upstream.\n\ntiocspgrp() takes two tty_struct pointers: One to the tty that userspace\npassed to ioctl() (`tty`) and one to the TTY being changed (`real_tty`).\nThese pointers are different when ioctl() is called with a master fd.\n\nTo properly lock real_tty->pgrp, we must take real_tty->ctrl_lock.\n\nThis bug makes it possible for racing ioctl(TIOCSPGRP, ...) calls on\nboth sides of a PTY pair to corrupt the refcount of `struct pid`,\nleading to use-after-free errors.\n\nFixes: 47f86834bbd4 (\"redo locking of tty->pgrp\")\nCC: stable@kernel.org\nSigned-off-by: Jann Horn \nReviewed-by: Jiri Slaby \nSigned-off-by: Greg Kroah-Hartman ","shortMessageHtmlLink":"tty: Fix ->pgrp locking in tiocspgrp()"}},{"before":null,"after":"76665b2f4e86e770eeac25afd466af482021556d","ref":"refs/heads/lineage-18.1","pushedAt":"2023-05-21T17:04:03.710Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"tty: Fix ->pgrp locking in tiocspgrp()\n\ncommit 54ffccbf053b5b6ca4f6e45094b942fab92a25fc upstream.\n\ntiocspgrp() takes two tty_struct pointers: One to the tty that userspace\npassed to ioctl() (`tty`) and one to the TTY being changed (`real_tty`).\nThese pointers are different when ioctl() is called with a master fd.\n\nTo properly lock real_tty->pgrp, we must take real_tty->ctrl_lock.\n\nThis bug makes it possible for racing ioctl(TIOCSPGRP, ...) calls on\nboth sides of a PTY pair to corrupt the refcount of `struct pid`,\nleading to use-after-free errors.\n\nFixes: 47f86834bbd4 (\"redo locking of tty->pgrp\")\nCC: stable@kernel.org\nSigned-off-by: Jann Horn \nReviewed-by: Jiri Slaby \nSigned-off-by: Greg Kroah-Hartman ","shortMessageHtmlLink":"tty: Fix ->pgrp locking in tiocspgrp()"}},{"before":null,"after":"76665b2f4e86e770eeac25afd466af482021556d","ref":"refs/heads/lineage-18.1","pushedAt":"2023-05-21T17:04:03.710Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"TARKZiM","name":"Jia Cheng She","path":"/TARKZiM","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6601824?s=80&v=4"},"commit":{"message":"tty: Fix ->pgrp locking in tiocspgrp()\n\ncommit 54ffccbf053b5b6ca4f6e45094b942fab92a25fc upstream.\n\ntiocspgrp() takes two tty_struct pointers: One to the tty that userspace\npassed to ioctl() (`tty`) and one to the TTY being changed (`real_tty`).\nThese pointers are different when ioctl() is called with a master fd.\n\nTo properly lock real_tty->pgrp, we must take real_tty->ctrl_lock.\n\nThis bug makes it possible for racing ioctl(TIOCSPGRP, ...) calls on\nboth sides of a PTY pair to corrupt the refcount of `struct pid`,\nleading to use-after-free errors.\n\nFixes: 47f86834bbd4 (\"redo locking of tty->pgrp\")\nCC: stable@kernel.org\nSigned-off-by: Jann Horn \nReviewed-by: Jiri Slaby \nSigned-off-by: Greg Kroah-Hartman ","shortMessageHtmlLink":"tty: Fix ->pgrp locking in tiocspgrp()"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"Y3Vyc29yOnYyOpK7MjAyNC0wOS0xOFQxODoxOTo0NC4wMDAwMDBazwAAAAS6B95y","startCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wOS0xOFQxODoxOTo0NC4wMDAwMDBazwAAAAS6B95y","endCursor":"Y3Vyc29yOnYyOpK7MjAyMy0wNS0yMVQxNzowNDowMy43MTA1NzlazwAAAAMxEgx5"}},"title":"Activity · TARKZiM/android_kernel_htc_msm8998"}